4.19.155-rc1 review
Subsystems (4)
Greg Kroah-Hartman
maintainercover 4.19.155-rc1 review
This is the start of the stable review cycle for the 4.19.155 release. There are 191 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
Responses should be made by Thu, 05 Nov 2020 20:29:58 +0000. Anything received after that time might be too late.
The whole patch series can be found in one patch at:
https://www.kernel.org/pub/linux/kernel/v4.x/stable-review/patch-4.19.155-rc1.gz
or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-4.19.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 4.19.155-rc1
Alexander Sverdlin <[email protected]>
staging: octeon: Drop on uncorrectable alignment or FCS error
Alexander Sverdlin <[email protected]>
staging: octeon: repair "fixed-link" support
Ian Abbott <[email protected]>
staging: comedi: cb_pcidas: Allow 2-channel commands for AO subdevice
Marc Zyngier <[email protected]>
KVM: arm64: Fix AArch32 handling of DBGD{CCINT,SCRext} and DBGVCR
Andy Shevchenko <[email protected]>
device property: Don't clear secondary pointer for shared primary firmware node
Andy Shevchenko <[email protected]>
device property: Keep secondary firmware node secondary by type
Krzysztof Kozlowski <[email protected]>
ARM: s3c24xx: fix missing system reset
Krzysztof Kozlowski <[email protected]>
ARM: samsung: fix PM debug build with DEBUG_LL but !MMU
Frank Wunderlich <[email protected]>
arm: dts: mt7623: add missing pause for switchport
Helge Deller <[email protected]>
hil/parisc: Disable HIL driver when it gets stuck
Matthew Wilcox (Oracle) <[email protected]>
cachefiles: Handle readpage error correctly
Jisheng Zhang <[email protected]>
arm64: berlin: Select DW_APB_TIMER_OF
Linus Torvalds <[email protected]>
tty: make FONTX ioctl use the tty pointer they were actually passed
Bartosz Golaszewski <[email protected]>
rtc: rx8010: don't modify the global rtc ops
Dave Airlie <[email protected]>
drm/ttm: fix eviction valuable range check.
Constantine Sapuntzakis <[email protected]>
ext4: fix superblock checksum calculation race
Luo Meng <[email protected]>
ext4: fix invalid inode checksum
Dinghao Liu <[email protected]>
ext4: fix error handling code in add_new_gdb
Eric Biggers <[email protected]>
ext4: fix leaking sysfs kobject after failed mount
Stefano Garzarella <[email protected]>
vringh: fix __vringh_iov() when riov and wiov are different
Qiujun Huang <[email protected]>
ring-buffer: Return 0 on success from ring_buffer_resize()
Matthew Wilcox (Oracle) <[email protected]>
9P: Cast to loff_t before multiplying
Ilya Dryomov <[email protected]>
libceph: clear con->out_msg on Policy::stateful_server faults
Matthew Wilcox (Oracle) <[email protected]>
ceph: promote to unsigned long long before shifting
Takashi Iwai <[email protected]>
drm/amd/display: Don't invoke kgdb_breakpoint() unconditionally
Madhav Chauhan <[email protected]>
drm/amdgpu: don't map BO in reserved region
Krzysztof Kozlowski <[email protected]>
i2c: imx: Fix external abort on interrupt in exit paths
Krzysztof Kozlowski <[email protected]>
ia64: fix build error with !COREDUMP
Zhihao Cheng <[email protected]>
ubi: check kthread_should_stop() after the setting of task state
Jiri Olsa <[email protected]>
perf python scripting: Fix printable strings in python3 scripts
Zhihao Cheng <[email protected]>
ubifs: dent: Fix some potential memory leaks while iterating entries
Chuck Lever <[email protected]>
NFSD: Add missing NFSv2 .pc_func methods
Olga Kornievskaia <[email protected]>
NFSv4.2: support EXCHGID4_FLAG_SUPP_FENCE_OPS 4.2 EXCHANGE_ID flag
Michael Neuling <[email protected]>
powerpc: Fix undetected data corruption with P9N DD2.1 VSX CI load emulation
Mahesh Salgaonkar <[email protected]>
powerpc/powernv/elog: Fix race while processing OPAL error log event.
Joel Stanley <[email protected]>
powerpc: Warn about use of smt_snooze_delay
Andrew Donnellan <[email protected]>
powerpc/rtas: Restrict RTAS requests from userspace
Sven Schnelle <[email protected]>
s390/stp: add locking to sysfs functions
Aneesh Kumar K.V <[email protected]>
powerpc/drmem: Make lmb_size 64 bit
Jonathan Cameron <[email protected]>
iio:gyro:itg3200: Fix timestamp alignment and prevent data leak.
Jonathan Cameron <[email protected]>
iio:adc:ti-adc12138 Fix alignment issue with timestamp
Jonathan Cameron <[email protected]>
iio:adc:ti-adc0832 Fix alignment issue with timestamp
Jonathan Cameron <[email protected]>
iio:light:si1145: Fix timestamp alignment and prevent data leak.
Paul Cercueil <[email protected]>
dmaengine: dma-jz4780: Fix race in jz4780_dma_tx_status
Jan Kara <[email protected]>
udf: Fix memory leak when mounting
Jason Gerecke <[email protected]>
HID: wacom: Avoid entering wacom_wac_pen_report for pad / battery
Jiri Slaby <[email protected]>
vt: keyboard, extend func_buf_lock to readers
Jiri Slaby <[email protected]>
vt: keyboard, simplify vt_kdgkbsent
Chris Wilson <[email protected]>
drm/i915: Force VT'd workarounds when running as a guest OS
Ran Wang <[email protected]>
usb: host: fsl-mph-dr-of: check return of dma_set_mask()
Li Jun <[email protected]>
usb: typec: tcpm: reset hard_reset_count for any disconnect
Jerome Brunet <[email protected]>
usb: cdc-acm: fix cooldown mechanism
Li Jun <[email protected]>
usb: dwc3: core: don't trigger runtime pm when remove driver
Li Jun <[email protected]>
usb: dwc3: core: add phy cleanup for probe error handling
Thinh Nguyen <[email protected]>
usb: dwc3: gadget: Check MPS of the request length
Thinh Nguyen <[email protected]>
usb: dwc3: ep0: Fix ZLP for OUT ep0 requests
Sandeep Singh <[email protected]>
usb: xhci: Workaround for S3 issue on AMD SNPS 3.0 xHC
Filipe Manana <[email protected]>
btrfs: fix use-after-free on readahead extent after failure to create it
Josef Bacik <[email protected]>
btrfs: cleanup cow block on error
Denis Efremov <[email protected]>
btrfs: use kvzalloc() to allocate clone_roots in btrfs_ioctl_send()
Filipe Manana <[email protected]>
btrfs: send, recompute reference path after orphanization of a directory
Filipe Manana <[email protected]>
btrfs: reschedule if necessary when logging directory items
Anand Jain <[email protected]>
btrfs: improve device scanning messages
Qu Wenruo <[email protected]>
btrfs: qgroup: fix wrong qgroup metadata reserve for delayed inode
Quinn Tran <[email protected]>
scsi: qla2xxx: Fix crash on session cleanup with unload
Helge Deller <[email protected]>
scsi: mptfusion: Fix null pointer dereferences in mptscsih_remove()
Martin Fuzzey <[email protected]>
w1: mxc_w1: Fix timeout resolution problem leading to bus error
Wei Huang <[email protected]>
acpi-cpufreq: Honor _PSD table setting on new AMD CPUs
Jamie Iles <[email protected]>
ACPI: debug: don't allow debugging when ACPI is disabled
Alex Hung <[email protected]>
ACPI: video: use ACPI backlight for HP 635 Notebook
Ben Hutchings <[email protected]>
ACPI / extlog: Check for RDMSR failure
[email protected] <[email protected]>
ACPI: button: fix handling lid state changes when input device closed
Ashish Sangwan <[email protected]>
NFS: fix nfs_path in case of a rename retry
Jan Kara <[email protected]>
fs: Don't invalidate page buffers in block_write_full_page()
Hans de Goede <[email protected]>
media: uvcvideo: Fix uvc_ctrl_fixup_xu_info() not having any effect
Marek Behún <[email protected]>
leds: bcm6328, bcm6358: use devres LED registering function
Kim Phillips <[email protected]>
perf/x86/amd/ibs: Fix raw sample data accumulation
Kim Phillips <[email protected]>
perf/x86/amd/ibs: Don't include randomized bits in get_ibs_op_count()
Raul E Rangel <[email protected]>
mmc: sdhci-acpi: AMDI0040: Set SDHCI_QUIRK2_PRESET_VALUE_BROKEN
Song Liu <[email protected]>
md/raid5: fix oops during stripe resizing
Chao Leng <[email protected]>
nvme-rdma: fix crash when connect rejected
Douglas Gilbert <[email protected]>
sgl_alloc_order: fix memory leak
Xiubo Li <[email protected]>
nbd: make the config put is called before the notifying the waiter
Krzysztof Kozlowski <[email protected]>
ARM: dts: s5pv210: remove dedicated 'audio-subsystem' node
Krzysztof Kozlowski <[email protected]>
ARM: dts: s5pv210: move PMU node out of clock controller
Krzysztof Kozlowski <[email protected]>
ARM: dts: s5pv210: remove DMA controller bus node name to fix dtschema warnings
Dan Carpenter <[email protected]>
memory: emif: Remove bogus debugfs error handling
Tony Lindgren <[email protected]>
ARM: dts: omap4: Fix sgx clock rate for 4430
Yoshihiro Shimoda <[email protected]>
arm64: dts: renesas: ulcb: add full-pwr-cycle-in-suspend into eMMC nodes
Ronnie Sahlberg <[email protected]>
cifs: handle -EINTR in cifs_setattr
Anant Thazhemadam <[email protected]>
gfs2: add validation checks for size of superblock
Jan Kara <[email protected]>
ext4: Detect already used quota file early
Madhuparna Bhowmik <[email protected]>
drivers: watchdog: rdc321x_wdt: Fix race condition bugs
Anant Thazhemadam <[email protected]>
net: 9p: initialize sun_server.sun_path to have addr's value only when addr is valid
Tero Kristo <[email protected]>
clk: ti: clockdomain: fix static checker warning
Chris Lew <[email protected]>
rpmsg: glink: Use complete_all for open states
Michael Chan <[email protected]>
bnxt_en: Log unknown link speed appropriately.
Zhao Heming <[email protected]>
md/bitmap: md_bitmap_get_counter returns wrong blocks
Anand Jain <[email protected]>
btrfs: fix replace of seed device
Fangzhi Zuo <[email protected]>
drm/amd/display: HDMI remote sink need mode validation for Linux
Xiongfeng Wang <[email protected]>
power: supply: test_power: add missing newlines when printing parameters by sysfs
Diana Craciun <[email protected]>
bus/fsl_mc: Do not rely on caller to provide non NULL mc_io
Xie He <[email protected]>
drivers/net/wan/hdlc_fr: Correctly handle special skb->protocol values
Jonathan Cameron <[email protected]>
ACPI: Add out of bounds and numa_off protections to pxm_to_node()
Darrick J. Wong <[email protected]>
xfs: don't free rt blocks when we're doing a REMAP bunmapi call
Zhengyuan Liu <[email protected]>
arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE
Peter Chen <[email protected]>
usb: xhci: omit duplicate actions when suspending a runtime suspended host.
Lang Dai <[email protected]>
uio: free uio id after uio file node is freed
Oliver Neukum <[email protected]>
USB: adutux: fix debugging
Alain Volmat <[email protected]>
cpufreq: sti-cpufreq: add stih418 support
Zong Li <[email protected]>
riscv: Define AT_VECTOR_SIZE_ARCH for ARCH_DLINFO
Daniel W. S. Almeida <[email protected]>
media: uvcvideo: Fix dereference of out-of-bound list iterator
Douglas Anderson <[email protected]>
kgdb: Make "kgdbcon" work properly with "kgdb_earlycon"
Masami Hiramatsu <[email protected]>
ia64: kprobes: Use generic kretprobe trampoline handler
John Ogness <[email protected]>
printk: reduce LOG_BUF_SHIFT range for H8300
Valentin Schneider <[email protected]>
arm64: topology: Stop using MPIDR for topology information
Antonio Borneo <[email protected]>
drm/bridge/synopsys: dsi: add support for non-continuous HS clock
Madhuparna Bhowmik <[email protected]>
mmc: via-sdmmc: Fix data race bug
Hans Verkuil <[email protected]>
media: imx274: fix frame interval handling
Tom Rix <[email protected]>
media: tw5864: check status of tw5864_frameinterval_get
Badhri Jagan Sridharan <[email protected]>
usb: typec: tcpm: During PR_SWAP, source caps should be sent only after tSwapSourceStart
Xia Jiang <[email protected]>
media: platform: Improve queue set up flow for bug fixing
Hans Verkuil <[email protected]>
media: videodev2.h: RGB BT2020 and HSV are always full range
Nadezda Lutovinova <[email protected]>
drm/brige/megachips: Add checking if ge_b850v3_lvds_init() is working correctly
Sathishkumar Muruganandam <[email protected]>
ath10k: fix VHT NSS calculation when STBC is enabled
Wen Gong <[email protected]>
ath10k: start recovery process when payload length exceeds max htc length for sdio
Tom Rix <[email protected]>
video: fbdev: pvr2fb: initialize variables
Darrick J. Wong <[email protected]>
xfs: fix realtime bitmap/summary file truncation when growing rt volume
Krzysztof Kozlowski <[email protected]>
power: supply: bq27xxx: report "not charging" on all types
Douglas Anderson <[email protected]>
ARM: 8997/2: hw_breakpoint: Handle inexact watchpoint addresses
Johannes Berg <[email protected]>
um: change sigio_spinlock to a mutex
Chao Yu <[email protected]>
f2fs: fix to check segment boundary during SIT page readahead
Chao Yu <[email protected]>
f2fs: fix uninit-value in f2fs_lookup
Zhang Qilong <[email protected]>
f2fs: add trace exit in exception path
Nicholas Piggin <[email protected]>
sparc64: remove mm_cpumask clearing to fix kthread_use_mm race
Nicholas Piggin <[email protected]>
powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Oliver O'Halloran <[email protected]>
powerpc/powernv/smp: Fix spurious DBG() warning
Mateusz Nosek <[email protected]>
futex: Fix incorrect should_fail_futex() handling
Sascha Hauer <[email protected]>
ata: sata_nv: Fix retrieving of active qcs
Alok Prasad <[email protected]>
RDMA/qedr: Fix memory leak in iWARP CM
Amit Cohen <[email protected]>
mlxsw: core: Fix use-after-free in mlxsw_emad_trans_finish()
Jiri Slaby <[email protected]>
x86/unwind/orc: Fix inactive tasks with stack pointer in %sp on GCC 10 compiled kernels
Juergen Gross <[email protected]>
xen/events: block rogue events for some time
Juergen Gross <[email protected]>
xen/events: defer eoi in case of excessive number of events
Juergen Gross <[email protected]>
xen/events: use a common cpu hotplug hook for event channels
Juergen Gross <[email protected]>
xen/events: switch user event channels to lateeoi model
Juergen Gross <[email protected]>
xen/pciback: use lateeoi irq binding
Juergen Gross <[email protected]>
xen/pvcallsback: use lateeoi irq binding
Juergen Gross <[email protected]>
xen/scsiback: use lateeoi irq binding
Juergen Gross <[email protected]>
xen/netback: use lateeoi irq binding
Juergen Gross <[email protected]>
xen/blkback: use lateeoi irq binding
Juergen Gross <[email protected]>
xen/events: add a new "late EOI" evtchn framework
Juergen Gross <[email protected]>
xen/events: fix race in evtchn_fifo_unmask()
Juergen Gross <[email protected]>
xen/events: add a proper barrier to 2-level uevent unmasking
Juergen Gross <[email protected]>
xen/events: avoid removing an event channel while handling it
Juergen Gross <[email protected]>
xen/events: don't use chip_data for legacy IRQs
Greg Kroah-Hartman <[email protected]>
Revert "block: ratelimit handle_bad_sector() message"
Eric Biggers <[email protected]>
fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext
Eric Biggers <[email protected]>
fscrypt: only set dentry_operations on ciphertext dentries
Eric Biggers <[email protected]>
fs, fscrypt: clear DCACHE_ENCRYPTED_NAME when unaliasing directory
Eric Biggers <[email protected]>
fscrypt: fix race allowing rename() and link() of ciphertext dentries
Eric Biggers <[email protected]>
fscrypt: clean up and improve dentry revalidation
Eric Biggers <[email protected]>
fscrypt: return -EXDEV for incompatible rename or link into encrypted dir
Geert Uytterhoeven <[email protected]>
ata: sata_rcar: Fix DMA boundary mask
Peter Zijlstra <[email protected]>
serial: pl011: Fix lockdep splat when handling magic-sysrq interrupt
Gustavo A. R. Silva <[email protected]>
mtd: lpddr: Fix bad logic in print_drs_error
Jason Gunthorpe <[email protected]>
RDMA/addr: Fix race with netevent_callback()/rdma_addr_cancel()
Frederic Barrat <[email protected]>
cxl: Rework error message for incompatible slots
Jia-Ju Bai <[email protected]>
p54: avoid accessing the data mapped to streaming DMA
Roberto Sassu <[email protected]>
evm: Check size of security.evm before using it
Song Liu <[email protected]>
bpf: Fix comment for helper bpf_current_task_under_cgroup()
Miklos Szeredi <[email protected]>
fuse: fix page dereference after free
Juergen Gross <[email protected]>
x86/xen: disable Firmware First mode for correctable memory errors
Kim Phillips <[email protected]>
arch/x86/amd/ibs: Fix re-arming IBS Fetch
Raju Rangoju <[email protected]>
cxgb4: set up filter action after rewrites
Heiner Kallweit <[email protected]>
r8169: fix issue with forced threading in combination with shared interrupts
Tung Nguyen <[email protected]>
tipc: fix memory leak caused by tipc_buf_append()
Arjun Roy <[email protected]>
tcp: Prevent low rmem stalls with SO_RCVLOWAT.
Andrew Gabbasov <[email protected]>
ravb: Fix bit fields checking in ravb_hwtstamp_get()
Aleksandr Nogikh <[email protected]>
netem: fix zero division in tabledist
Ido Schimmel <[email protected]>
mlxsw: core: Fix memory leak on module removal
Masahiro Fujiwara <[email protected]>
gtp: fix an use-before-init in gtp_newlink()
Vinay Kumar Yadav <[email protected]>
chelsio/chtls: fix tls record info to user
Vinay Kumar Yadav <[email protected]>
chelsio/chtls: fix memory leaks in CPL handlers
Vinay Kumar Yadav <[email protected]>
chelsio/chtls: fix deadlock issue
Michael Schaller <[email protected]>
efivarfs: Replace invalid slashes with exclamation marks in dentries.
Randy Dunlap <[email protected]>
x86/PCI: Fix intel_mid_pci.c build error when ACPI is not enabled
Nick Desaulniers <[email protected]>
arm64: link with -z norelro regardless of CONFIG_RELOCATABLE
Marc Zyngier <[email protected]>
arm64: Run ARCH_WORKAROUND_1 enabling code on all CPUs
Rasmus Villemoes <[email protected]>
scripts/setlocalversion: make git describe output more reliable
Josh Poimboeuf <[email protected]>
objtool: Support Clang non-section symbols in ORC generation
-------------
Diffstat:
Documentation/admin-guide/kernel-parameters.txt | 8 + Documentation/filesystems/fscrypt.rst | 12 +- Documentation/media/uapi/v4l/colorspaces-defs.rst | 9 +- .../media/uapi/v4l/colorspaces-details.rst | 5 +- Makefile | 4 +- arch/arm/Kconfig | 2 + arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts | 1 + arch/arm/boot/dts/omap4.dtsi | 2 +- arch/arm/boot/dts/omap443x.dtsi | 10 + arch/arm/boot/dts/s5pv210.dtsi | 127 +++--- arch/arm/kernel/hw_breakpoint.c | 100 +++-- arch/arm/plat-samsung/Kconfig | 1 + arch/arm64/Kconfig.platforms | 1 + arch/arm64/Makefile | 4 +- arch/arm64/boot/dts/renesas/ulcb.dtsi | 1 + arch/arm64/include/asm/kvm_host.h | 1 + arch/arm64/include/asm/numa.h | 3 + arch/arm64/kernel/cpu_errata.c | 8 + arch/arm64/kernel/topology.c | 32 +- arch/arm64/kvm/sys_regs.c | 6 +- arch/arm64/mm/numa.c | 6 +- arch/ia64/kernel/Makefile | 2 +- arch/ia64/kernel/kprobes.c | 77 +--- arch/powerpc/Kconfig | 14 + arch/powerpc/include/asm/drmem.h | 4 +- arch/powerpc/include/asm/mmu_context.h | 2 +- arch/powerpc/kernel/rtas.c | 153 +++++++ arch/powerpc/kernel/sysfs.c | 42 +- arch/powerpc/kernel/traps.c | 2 +- arch/powerpc/platforms/powernv/opal-elog.c | 33 +- arch/powerpc/platforms/powernv/smp.c | 2 +- arch/riscv/include/uapi/asm/auxvec.h | 3 + arch/s390/kernel/time.c | 118 ++++-- arch/sparc/kernel/smp_64.c | 65 +-- arch/um/kernel/sigio.c | 6 +- arch/x86/events/amd/ibs.c | 53 ++- arch/x86/include/asm/msr-index.h | 1 + arch/x86/kernel/unwind_orc.c | 9 +- arch/x86/pci/intel_mid_pci.c | 1 + arch/x86/xen/enlighten_pv.c | 9 + block/blk-core.c | 9 +- drivers/acpi/acpi_dbg.c | 3 + drivers/acpi/acpi_extlog.c | 6 +- drivers/acpi/button.c | 13 +- drivers/acpi/numa.c | 2 +- drivers/acpi/video_detect.c | 9 + drivers/ata/sata_nv.c | 2 +- drivers/ata/sata_rcar.c | 2 +- drivers/base/core.c | 4 +- drivers/block/nbd.c | 2 +- drivers/block/xen-blkback/blkback.c | 22 +- drivers/block/xen-blkback/xenbus.c | 5 +- drivers/bus/fsl-mc/mc-io.c | 7 +- drivers/clk/ti/clockdomain.c | 2 + drivers/cpufreq/acpi-cpufreq.c | 3 +- drivers/cpufreq/sti-cpufreq.c | 6 +- drivers/crypto/chelsio/chtls/chtls_cm.c | 29 +- drivers/crypto/chelsio/chtls/chtls_io.c | 7 +- drivers/dma/dma-jz4780.c | 7 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 10 + drivers/gpu/drm/amd/display/dc/core/dc_link.c | 2 +- drivers/gpu/drm/amd/display/dc/os_types.h | 2 +- .../drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 12 +- drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c | 9 +- drivers/gpu/drm/i915/i915_drv.h | 6 +- drivers/gpu/drm/ttm/ttm_bo.c | 2 +- drivers/hid/wacom_wac.c | 4 +- drivers/i2c/busses/i2c-imx.c | 24 +- drivers/iio/adc/ti-adc0832.c | 11 +- drivers/iio/adc/ti-adc12138.c | 13 +- drivers/iio/gyro/itg3200_buffer.c | 15 +- drivers/iio/light/si1145.c | 19 +- drivers/infiniband/core/addr.c | 11 +- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 1 + drivers/input/serio/hil_mlc.c | 21 +- drivers/input/serio/hp_sdc_mlc.c | 8 +- drivers/leds/leds-bcm6328.c | 2 +- drivers/leds/leds-bcm6358.c | 2 +- drivers/md/md-bitmap.c | 2 +- drivers/md/raid5.c | 4 +- drivers/media/i2c/imx274.c | 8 +- drivers/media/pci/tw5864/tw5864-video.c | 6 + drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c | 7 + drivers/media/usb/uvc/uvc_ctrl.c | 27 +- drivers/memory/emif.c | 33 +- drivers/message/fusion/mptscsih.c | 13 +- drivers/misc/cxl/pci.c | 4 +- drivers/mmc/host/sdhci-acpi.c | 37 ++ drivers/mmc/host/via-sdmmc.c | 3 + drivers/mtd/ubi/wl.c | 13 + drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +- drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 56 ++- drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h | 4 + drivers/net/ethernet/mellanox/mlxsw/core.c | 5 + drivers/net/ethernet/realtek/r8169.c | 4 +- drivers/net/ethernet/renesas/ravb_main.c | 10 +- drivers/net/gtp.c | 16 +- drivers/net/wan/hdlc_fr.c | 98 ++--- drivers/net/wireless/ath/ath10k/htt_rx.c | 8 +- drivers/net/wireless/ath/ath10k/sdio.c | 4 + drivers/net/wireless/intersil/p54/p54pci.c | 4 +- drivers/net/xen-netback/common.h | 15 + drivers/net/xen-netback/interface.c | 61 ++- drivers/net/xen-netback/netback.c | 11 +- drivers/net/xen-netback/rx.c | 13 +- drivers/nvme/host/rdma.c | 1 - drivers/power/supply/bq27xxx_battery.c | 6 +- drivers/power/supply/test_power.c | 6 + drivers/rpmsg/qcom_glink_native.c | 6 +- drivers/rtc/rtc-rx8010.c | 24 +- drivers/scsi/qla2xxx/qla_target.c | 13 +- drivers/staging/comedi/drivers/cb_pcidas.c | 1 + drivers/staging/octeon/ethernet-mdio.c | 6 - drivers/staging/octeon/ethernet-rx.c | 34 +- drivers/staging/octeon/ethernet.c | 9 + drivers/tty/serial/amba-pl011.c | 11 +- drivers/tty/vt/keyboard.c | 39 +- drivers/tty/vt/vt_ioctl.c | 32 +- drivers/uio/uio.c | 4 +- drivers/usb/class/cdc-acm.c | 12 +- drivers/usb/class/cdc-acm.h | 3 +- drivers/usb/dwc3/core.c | 15 +- drivers/usb/dwc3/ep0.c | 11 +- drivers/usb/dwc3/gadget.c | 4 +- drivers/usb/host/fsl-mph-dr-of.c | 9 +- drivers/usb/host/xhci-pci.c | 17 + drivers/usb/host/xhci.c | 7 +- drivers/usb/host/xhci.h | 1 + drivers/usb/misc/adutux.c | 1 + drivers/usb/typec/tcpm.c | 8 +- drivers/vhost/vringh.c | 9 +- drivers/video/fbdev/pvr2fb.c | 2 + drivers/w1/masters/mxc_w1.c | 14 +- drivers/watchdog/rdc321x_wdt.c | 5 +- drivers/xen/events/events_2l.c | 9 +- drivers/xen/events/events_base.c | 451 +++++++++++++++++++-- drivers/xen/events/events_fifo.c | 83 ++-- drivers/xen/events/events_internal.h | 20 +- drivers/xen/evtchn.c | 7 +- drivers/xen/pvcalls-back.c | 76 ++-- drivers/xen/xen-pciback/pci_stub.c | 14 +- drivers/xen/xen-pciback/pciback.h | 12 +- drivers/xen/xen-pciback/pciback_ops.c | 48 ++- drivers/xen/xen-pciback/xenbus.c | 2 +- drivers/xen/xen-scsiback.c | 23 +- fs/9p/vfs_file.c | 4 +- fs/btrfs/ctree.c | 6 + fs/btrfs/delayed-inode.c | 3 +- fs/btrfs/dev-replace.c | 2 +- fs/btrfs/reada.c | 2 + fs/btrfs/send.c | 74 +++- fs/btrfs/tree-log.c | 8 + fs/btrfs/volumes.c | 14 +- fs/buffer.c | 16 - fs/cachefiles/rdwr.c | 3 +- fs/ceph/addr.c | 2 +- fs/cifs/inode.c | 13 +- fs/crypto/crypto.c | 58 +-- fs/crypto/fname.c | 1 + fs/crypto/hooks.c | 34 +- fs/crypto/policy.c | 3 +- fs/dcache.c | 15 + fs/efivarfs/super.c | 3 + fs/ext4/ext4.h | 62 ++- fs/ext4/inode.c | 11 +- fs/ext4/namei.c | 76 ++-- fs/ext4/resize.c | 4 +- fs/ext4/super.c | 17 + fs/f2fs/checkpoint.c | 8 +- fs/f2fs/dir.c | 8 +- fs/f2fs/namei.c | 17 +- fs/fuse/dev.c | 28 +- fs/gfs2/ops_fstype.c | 18 +- fs/nfs/namespace.c | 12 +- fs/nfs/nfs4proc.c | 9 +- fs/nfsd/nfsproc.c | 16 + fs/ubifs/debug.c | 1 + fs/ubifs/dir.c | 8 +- fs/udf/super.c | 21 +- fs/xfs/libxfs/xfs_bmap.c | 19 +- fs/xfs/xfs_rtalloc.c | 10 +- include/linux/dcache.h | 2 +- include/linux/fscrypt.h | 34 +- include/linux/fscrypt_notsupp.h | 9 +- include/linux/fscrypt_supp.h | 6 +- include/linux/hil_mlc.h | 2 +- include/linux/mtd/pfow.h | 2 +- include/linux/usb/pd.h | 1 + include/uapi/linux/bpf.h | 4 +- include/uapi/linux/nfs4.h | 3 + include/uapi/linux/videodev2.h | 17 +- include/xen/events.h | 29 +- init/Kconfig | 3 +- kernel/debug/debug_core.c | 22 +- kernel/futex.c | 4 +- kernel/trace/ring_buffer.c | 8 +- lib/scatterlist.c | 2 +- net/9p/trans_fd.c | 2 +- net/ceph/messenger.c | 5 + net/ipv4/tcp.c | 2 + net/ipv4/tcp_input.c | 3 +- net/sched/sch_netem.c | 9 +- net/tipc/msg.c | 5 +- scripts/setlocalversion | 21 +- security/integrity/evm/evm_main.c | 6 + tools/include/uapi/linux/bpf.h | 4 +- tools/objtool/orc_gen.c | 33 +- tools/perf/util/print_binary.c | 2 +- 208 files changed, 2480 insertions(+), 1146 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/191 objtool: Support Clang non-section symbols in ORC generation
From: Josh Poimboeuf <[email protected]>
commit e81e0724432542af8d8c702c31e9d82f57b1ff31 upstream.
When compiling the kernel with AS=clang, objtool produces a lot of warnings:
warning: objtool: missing symbol for section .text
warning: objtool: missing symbol for section .init.text
warning: objtool: missing symbol for section .ref.text
It then fails to generate the ORC table.
The problem is that objtool assumes text section symbols always exist. But the Clang assembler is aggressive about removing them.
When generating relocations for the ORC table, objtool always tries to reference instructions by their section symbol offset. If the section symbol doesn't exist, it bails.
Do a fallback: when a section symbol isn't available, reference a function symbol instead.
--- tools/objtool/orc_gen.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-)
tools/objtool/orc_gen.c +26 -7
@@ -100,11 +100,6 @@ static int create_orc_entry(struct sectistruct orc_entry *orc;struct rela *rela; -if (!insn_sec->sym) { -WARN ("missing symbol for section %s" , insn_sec->name); -return -1 ; - } -/* populate ORC data */ orc = (struct orc_entry *)u_sec->data->d_buf + idx;memcpy (orc, o,sizeof (*orc));
@@ -117,8 +112,32 @@ static int create_orc_entry(struct secti }memset (rela,0 ,sizeof (*rela)); - rela->sym = insn_sec->sym; - rela->addend = insn_off; +if (insn_sec->sym) { + rela->sym = insn_sec->sym; + rela->addend = insn_off; + }else { +/* +* The Clang assembler doesn't produce section symbols, so we +* have to reference the function symbol instead: +*/ + rela->sym =find_symbol_containing (insn_sec, insn_off); +if (!rela->sym) { +/* +* Hack alert. This happens when we need to reference +* the NOP pad insn immediately after the function. +*/ + rela->sym =find_symbol_containing (insn_sec, + insn_off -1 ); + } +if (!rela->sym) { +WARN ("missing symbol for insn at offset 0x%lx\n" , + insn_off); +return -1 ; + } + + rela->addend = insn_off - rela->sym->offset; + } + rela->type =R_X86_64_PC32 ; rela->offset = idx *sizeof (int );
Greg Kroah-Hartman
maintainerPATCH 2/191 scripts/setlocalversion: make git describe output more reliable
In reply to Greg Kroah-HartmanFrom: Rasmus Villemoes <[email protected]>
commit 548b8b5168c90c42e88f70fcf041b4ce0b8e7aa8 upstream.
When building for an embedded target using Yocto, we're sometimes observing that the version string that gets built into vmlinux (and thus what uname -a reports) differs from the path under /lib/modules/ where modules get installed in the rootfs, but only in the length of the -gabc123def suffix. Hence modprobe always fails.
The problem is that Yocto has the concept of "sstate" (shared state), which allows different developers/buildbots/etc. to share build artifacts, based on a hash of all the metadata that went into building that artifact - and that metadata includes all dependencies (e.g. the compiler used etc.). That normally works quite well; usually a clean build (without using any sstate cache) done by one developer ends up being binary identical to a build done on another host. However, one thing that can cause two developers to end up with different builds [and thus make one's vmlinux package incompatible with the other's kernel-dev package], which is not captured by the metadata hashing, is this git describe: The output of that can be affected by
(1) git version: before 2.11 git defaulted to a minimum of 7, since 2.11 (git.git commit e6c587) the default is dynamic based on the number of objects in the repo
(2) hence even if both run the same git version, the output can differ based on how many remotes are being tracked (or just lots of local development branches or plain old garbage)
(3) and of course somebody could have a core.abbrev config setting in ~/.gitconfig
So in order to avoid uname -a output relying on such random details of the build environment which are rather hard to ensure are consistent between developers and buildbots, make sure the abbreviated sha1 always consists of exactly 12 hex characters. That is consistent with the current rule for -stable patches, and is almost always enough to identify the head commit unambigously - in the few cases where it does not, the v5.4.3-00021- prefix would certainly nail it down.
--- scripts/setlocalversion | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-)
scripts/setlocalversion +15 -4
@@ -45,7 +45,7 @@ scm_version() # Check for git and a git repo. if test -z "$(git rev-parse --show-cdup 2>/dev/null)" && - head=`git rev-parse --verify --short HEAD 2>/dev/null`; then + head=$(git rev-parse --verify HEAD 2>/dev/null); then # If we are at a tagged commit (like "v2.6.30-rc6"), we ignore # it, because this version is defined in the top level Makefile.
@@ -59,11 +59,22 @@ scm_version() fi # If we are past a tagged commit (like # "v2.6.30-rc5-302-g72357d5"), we pretty print it. - if atag="`git describe 2>/dev/null`"; then - echo "$atag" | awk -F- '{printf("-%05d-%s", $(NF-1),$(NF))}' + # + # Ensure the abbreviated sha1 has exactly 12 + # hex characters, to make the output + # independent of git version, local + # core.abbrev settings and/or total number of + # objects in the current repository - passing + # --abbrev=12 ensures a minimum of 12, and the + # awk substr() then picks the 'g' and first 12 + # hex chars. + if atag="$(git describe --abbrev=12 2>/dev/null)"; then + echo "$atag" | awk -F- '{printf("-%05d-%s", $(NF-1),substr($(NF),0,13))}' - # If we don't have a tag at all we print -g{commitish}. + # If we don't have a tag at all we print -g{commitish}, + # again using exactly 12 hex chars. else + head="$(echo $head | cut -c1-12)" printf '%s%s' -g $head fi fi
Greg Kroah-Hartman
maintainerPATCH 3/191 arm64: Run ARCH_WORKAROUND_1 enabling code on all CPUs
In reply to Greg Kroah-HartmanFrom: Marc Zyngier <[email protected]>
commit 18fce56134c987e5b4eceddafdbe4b00c07e2ae1 upstream.
Commit 73f381660959 ("arm64: Advertise mitigation of Spectre-v2, or lack thereof") changed the way we deal with ARCH_WORKAROUND_1, by moving most of the enabling code to the .matches() callback.
This has the unfortunate effect that the workaround gets only enabled on the first affected CPU, and no other.
In order to address this, forcefully call the .matches() callback from a .cpu_enable() callback, which brings us back to the original behaviour.
--- arch/arm64/kernel/cpu_errata.c | 8 ++++++++ 1 file changed, 8 insertions(+)
arch/arm64/kernel/cpu_errata.c +8 -0
@@ -619,6 +619,12 @@ check_branch_predictor(const struct arm6return (need_wa >0 ); } +static void +cpu_enable_branch_predictor_hardening (const struct arm64_cpu_capabilities *cap) +{ + cap->matches (cap,SCOPE_LOCAL_CPU ); +} +static const __maybe_unused struct midr_range tx2_family_cpus[] = {MIDR_ALL_VERSIONS (MIDR_BRCM_VULCAN ),MIDR_ALL_VERSIONS (MIDR_CAVIUM_THUNDERX2 ),
@@ -813,9 +819,11 @@ const struct arm64_cpu_capabilities arm6 },#endif { + .desc = "Branch predictor hardening", .capability = ARM64_HARDEN_BRANCH_PREDICTOR, .type = ARM64_CPUCAP_LOCAL_CPU_ERRATUM, .matches = check_branch_predictor, + .cpu_enable = cpu_enable_branch_predictor_hardening, }, #ifdef CONFIG_HARDEN_EL2_VECTORS {
Greg Kroah-Hartman
maintainerPATCH 4/191 arm64: link with -z norelro regardless of CONFIG_RELOCATABLE
In reply to Greg Kroah-HartmanFrom: Nick Desaulniers <[email protected]>
commit 3b92fa7485eba16b05166fddf38ab42f2ff6ab95 upstream.
With CONFIG_EXPERT=y, CONFIG_KASAN=y, CONFIG_RANDOMIZE_BASE=n, CONFIG_RELOCATABLE=n, we observe the following failure when trying to link the kernel image with LD=ld.lld:
error: section: .exit.data is not contiguous with other relro sections
ld.lld defaults to -z relro while ld.bfd defaults to -z norelro. This was previously fixed, but only for CONFIG_RELOCATABLE=y.
--- arch/arm64/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/Makefile +2 -2
@@ -10,7 +10,7 @@ # # Copyright (C) 1995-2001 by Russell King -LDFLAGS_vmlinux :=--no-undefined -X +LDFLAGS_vmlinux :=--no-undefined -X -z norelro CPPFLAGS_vmlinux.lds = -DTEXT_OFFSET=$(TEXT_OFFSET) GZFLAGS :=-9
@@ -18,7 +18,7 @@ ifeq ($(CONFIG_RELOCATABLE), y) # Pass --no-apply-dynamic-relocs to restore pre-binutils-2.27 behaviour # for relative relocs, since this leads to better Image compression # with the relocation offsets always being zero. -LDFLAGS_vmlinux += -shared -Bsymbolic -z notext -z norelro \ +LDFLAGS_vmlinux += -shared -Bsymbolic -z notext \ $(call ld-option, --no-apply-dynamic-relocs) endif
Greg Kroah-Hartman
maintainerPATCH 5/191 x86/PCI: Fix intel_mid_pci.c build error when ACPI is not enabled
In reply to Greg Kroah-HartmanFrom: Randy Dunlap <[email protected]>
commit 035fff1f7aab43e420e0098f0854470a5286fb83 upstream.
Fix build error when CONFIG_ACPI is not set/enabled by adding the header file <asm/acpi.h> which contains a stub for the function in the build error.
../arch/x86/pci/intel_mid_pci.c: In function ‘intel_mid_pci_init’:
../arch/x86/pci/intel_mid_pci.c:303:2: error: implicit declaration of function ‘acpi_noirq_set’; did you mean ‘acpi_irq_get’? [-Werror=implicit-function-declaration]
acpi_noirq_set();
--- arch/x86/pci/intel_mid_pci.c | 1 + 1 file changed, 1 insertion(+)
arch/x86/pci/intel_mid_pci.c +1 -0
@@ -33,6 +33,7 @@#include <asm/hw_irq.h> #include <asm/io_apic.h> #include <asm/intel-mid.h> +#include <asm/acpi.h> #define PCIE_CAP_OFFSET 0x100
Greg Kroah-Hartman
maintainerPATCH 6/191 efivarfs: Replace invalid slashes with exclamation marks in dentries.
In reply to Greg Kroah-HartmanFrom: Michael Schaller <[email protected]>
commit 336af6a4686d885a067ecea8c3c3dd129ba4fc75 upstream.
Without this patch efivarfs_alloc_dentry creates dentries with slashes in their name if the respective EFI variable has slashes in its name. This in turn causes EIO on getdents64, which prevents a complete directory listing of /sys/firmware/efi/efivars/.
This patch replaces the invalid shlashes with exclamation marks like kobject_set_name_vargs does for /sys/firmware/efi/vars/ to have consistently named dentries under /sys/firmware/efi/vars/ and /sys/firmware/efi/efivars/.
--- fs/efivarfs/super.c | 3 +++ 1 file changed, 3 insertions(+)
fs/efivarfs/super.c +3 -0
@@ -145,6 +145,9 @@ static int efivarfs_callback(efi_char16_ name[len +EFI_VARIABLE_GUID_LEN +1 ] ='\0' ; +/* replace invalid slashes like kobject_set_name_vargs does for /sys/firmware/efi/vars. */ +strreplace (name,'/' ,'!' ); + inode =efivarfs_get_inode (sb,d_inode (root),S_IFREG |0644 ,0 , is_removable);if (!inode)
Greg Kroah-Hartman
maintainerPATCH 7/191 chelsio/chtls: fix deadlock issue
In reply to Greg Kroah-HartmanFrom: Vinay Kumar Yadav <[email protected]>
[ Upstream commit 28e9dcd9172028263c8225c15c4e329e08475e89 ]
In chtls_pass_establish() we hold child socket lock using bh_lock_sock and we are again trying bh_lock_sock in add_to_reap_list, causing deadlock. Remove bh_lock_sock in add_to_reap_list() as lock is already held.
--- drivers/crypto/chelsio/chtls/chtls_cm.c | 2 -- 1 file changed, 2 deletions(-)
drivers/crypto/chelsio/chtls/chtls_cm.c +0 -2
@@ -1348,7 +1348,6 @@ static void add_to_reap_list(struct sockstruct chtls_sock *csk = sk->sk_user_data;local_bh_disable (); -bh_lock_sock (sk);release_tcp_port (sk);/* release the port immediately */ spin_lock (&reap_list_lock);
@@ -1357,7 +1356,6 @@ static void add_to_reap_list(struct sockif (!csk->passive_reap_next)schedule_work (&reap_task);spin_unlock (&reap_list_lock); -bh_unlock_sock (sk);local_bh_enable (); }
Greg Kroah-Hartman
maintainerPATCH 8/191 chelsio/chtls: fix memory leaks in CPL handlers
In reply to Greg Kroah-HartmanFrom: Vinay Kumar Yadav <[email protected]>
[ Upstream commit 6daa1da4e262b0cd52ef0acc1989ff22b5540264 ]
CPL handler functions chtls_pass_open_rpl() and chtls_close_listsrv_rpl() should return CPL_RET_BUF_DONE so that caller function will do skb free to avoid leak.
--- drivers/crypto/chelsio/chtls/chtls_cm.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-)
drivers/crypto/chelsio/chtls/chtls_cm.c +12 -15
@@ -696,14 +696,13 @@ static int chtls_pass_open_rpl(struct chif (rpl->status !=CPL_ERR_NONE ) {pr_info ("Unexpected PASS_OPEN_RPL status %u for STID %u\n" , rpl->status, stid); -return CPL_RET_BUF_DONE ; + }else { +cxgb4_free_stid (cdev->tids, stid, listen_ctx->lsk->sk_family); +sock_put (listen_ctx->lsk); +kfree (listen_ctx); +module_put (THIS_MODULE ); } -cxgb4_free_stid (cdev->tids, stid, listen_ctx->lsk->sk_family); -sock_put (listen_ctx->lsk); -kfree (listen_ctx); -module_put (THIS_MODULE ); - -return 0 ; +return CPL_RET_BUF_DONE ; }static int chtls_close_listsrv_rpl (struct chtls_dev *cdev,struct sk_buff *skb)
@@ -720,15 +719,13 @@ static int chtls_close_listsrv_rpl(strucif (rpl->status !=CPL_ERR_NONE ) {pr_info ("Unexpected CLOSE_LISTSRV_RPL status %u for STID %u\n" , rpl->status, stid); -return CPL_RET_BUF_DONE ; + }else { +cxgb4_free_stid (cdev->tids, stid, listen_ctx->lsk->sk_family); +sock_put (listen_ctx->lsk); +kfree (listen_ctx); +module_put (THIS_MODULE ); } - -cxgb4_free_stid (cdev->tids, stid, listen_ctx->lsk->sk_family); -sock_put (listen_ctx->lsk); -kfree (listen_ctx); -module_put (THIS_MODULE ); - -return 0 ; +return CPL_RET_BUF_DONE ; }static void chtls_purge_wr_queue (struct sock *sk)
Greg Kroah-Hartman
maintainerPATCH 9/191 chelsio/chtls: fix tls record info to user
In reply to Greg Kroah-HartmanFrom: Vinay Kumar Yadav <[email protected]>
[ Upstream commit 4f3391ce8f5a69e7e6d66d0a3fc654eb6dbdc919 ]
chtls_pt_recvmsg() receives a skb with tls header and subsequent skb with data, need to finalize the data copy whenever next skb with tls header is available. but here current tls header is overwritten by next available tls header, ends up corrupting user buffer data. fixing it by finalizing current record whenever next skb contains tls header.
v1->v2:
- Improved commit message.
--- drivers/crypto/chelsio/chtls/chtls_io.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/crypto/chelsio/chtls/chtls_io.c +5 -2
@@ -1549,6 +1549,7 @@ skip_copy: tp->urg_data =0 ;if ((avail + offset) >= skb->len) { +struct sk_buff *next_skb;if (ULP_SKB_CB (skb)->flags &ULPCB_FLAG_TLS_HDR ) { tp->copied_seq += skb->len; hws->rcvpld = skb->hdr_len;
@@ -1558,8 +1559,10 @@ skip_copy:chtls_free_skb (sk, skb); buffers_freed++; hws->copied_seq =0 ; -if (copied >= target && - !skb_peek (&sk->sk_receive_queue)) + next_skb =skb_peek (&sk->sk_receive_queue); +if (copied >= target && !next_skb) +break ; +if (ULP_SKB_CB (next_skb)->flags &ULPCB_FLAG_TLS_HDR )break ; } }while (len >0 );
Greg Kroah-Hartman
maintainerPATCH 10/191 gtp: fix an use-before-init in gtp_newlink()
In reply to Greg Kroah-HartmanFrom: Masahiro Fujiwara <[email protected]>
[ Upstream commit 51467431200b91682b89d31317e35dcbca1469ce ]
*_pdp_find() from gtp_encap_recv() would trigger a crash when a peer sends GTP packets while creating new GTP device.
RIP: 0010:gtp1_pdp_find.isra.0+0x68/0x90 [gtp] <SNIP> Call Trace: <IRQ> gtp_encap_recv+0xc2/0x2e0 [gtp] ? gtp1_pdp_find.isra.0+0x90/0x90 [gtp] udp_queue_rcv_one_skb+0x1fe/0x530 udp_queue_rcv_skb+0x40/0x1b0 udp_unicast_rcv_skb.isra.0+0x78/0x90 __udp4_lib_rcv+0x5af/0xc70 udp_rcv+0x1a/0x20 ip_protocol_deliver_rcu+0xc5/0x1b0 ip_local_deliver_finish+0x48/0x50 ip_local_deliver+0xe5/0xf0 ? ip_protocol_deliver_rcu+0x1b0/0x1b0
gtp_encap_enable() should be called after gtp_hastable_new() otherwise *_pdp_find() will access the uninitialized hash table.
--- drivers/net/gtp.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/net/gtp.c +8 -8
@@ -667,10 +667,6 @@ static int gtp_newlink(struct net *src_n gtp =netdev_priv (dev); - err =gtp_encap_enable (gtp, data); -if (err <0 ) -return err; -if (!data[IFLA_GTP_PDP_HASHSIZE ]) { hashsize =1024 ; }else {
@@ -681,12 +677,16 @@ static int gtp_newlink(struct net *src_n err =gtp_hashtable_new (gtp, hashsize);if (err <0 ) - goto out_encap; +return err; + + err =gtp_encap_enable (gtp, data); +if (err <0 ) + goto out_hashtable; err =register_netdevice (dev);if (err <0 ) {netdev_dbg (dev,"failed to register new netdev %d\n" , err); - goto out_hashtable; + goto out_encap; } gn =net_generic (dev_net (dev), gtp_net_id);
@@ -697,11 +697,11 @@ static int gtp_newlink(struct net *src_nreturn 0 ; +out_encap: +gtp_encap_disable (gtp); out_hashtable:kfree (gtp->addr_hash);kfree (gtp->tid_hash); -out_encap: -gtp_encap_disable (gtp);return err; }
Greg Kroah-Hartman
maintainerPATCH 11/191 mlxsw: core: Fix memory leak on module removal
In reply to Greg Kroah-HartmanFrom: Ido Schimmel <[email protected]>
[ Upstream commit adc80b6cfedff6dad8b93d46a5ea2775fd5af9ec ]
Free the devlink instance during the teardown sequence in the non-reload case to avoid the following memory leak.
unreferenced object 0xffff888232895000 (size 2048):
comm "modprobe", pid 1073, jiffies 4295568857 (age 164.871s) hex dump (first 32 bytes):
00 01 00 00 00 00 ad de 22 01 00 00 00 00 ad de ........".......
10 50 89 32 82 88 ff ff 10 50 89 32 82 88 ff ff .P.2.....P.2....
backtrace:
[<00000000c704e9a6>] __kmalloc+0x13a/0x2a0
[<00000000ee30129d>] devlink_alloc+0xff/0x760
[<0000000092ab3e5d>] 0xffffffffa042e5b0
[<000000004f3f8a31>] 0xffffffffa042f6ad
[<0000000092800b4b>] 0xffffffffa0491df3
[<00000000c4843903>] local_pci_probe+0xcb/0x170
[<000000006993ded7>] pci_device_probe+0x2c2/0x4e0
[<00000000a8e0de75>] really_probe+0x2c5/0xf90
[<00000000d42ba75d>] driver_probe_device+0x1eb/0x340
[<00000000bcc95e05>] device_driver_attach+0x294/0x300
[<000000000e2bc177>] __driver_attach+0x167/0x2f0
[<000000007d44cd6e>] bus_for_each_dev+0x148/0x1f0
[<000000003cd5a91e>] driver_attach+0x45/0x60
[<000000000041ce51>] bus_add_driver+0x3b8/0x720
[<00000000f5215476>] driver_register+0x230/0x4e0
[<00000000d79356f5>] __pci_register_driver+0x190/0x200--- drivers/net/ethernet/mellanox/mlxsw/core.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/mellanox/mlxsw/core.c +2 -0
@@ -1111,6 +1111,8 @@ void mlxsw_core_bus_device_unregister(stif (!reload)devlink_resources_unregister (devlink,NULL ); mlxsw_core->bus->fini (mlxsw_core->bus_priv); +if (!reload) +devlink_free (devlink);return ;
Greg Kroah-Hartman
maintainerPATCH 12/191 netem: fix zero division in tabledist
In reply to Greg Kroah-HartmanFrom: Aleksandr Nogikh <[email protected]>
[ Upstream commit eadd1befdd778a1eca57fad058782bd22b4db804 ]
Currently it is possible to craft a special netlink RTM_NEWQDISC command that can result in jitter being equal to 0x80000000. It is enough to set the 32 bit jitter to 0x02000000 (it will later be multiplied by 2^6) or just set the 64 bit jitter via TCA_NETEM_JITTER64. This causes an overflow during the generation of uniformly distributed numbers in tabledist(), which in turn leads to division by zero (sigma != 0, but sigma * 2 is 0).
The related fragment of code needs 32-bit division - see commit 9b0ed89 ("netem: remove unnecessary 64 bit modulus"), so switching to 64 bit is not an option.
Fix the issue by keeping the value of jitter within the range that can be adequately handled by tabledist() - [0;INT_MAX]. As negative std deviation makes no sense, take the absolute value of the passed value and cap it at INT_MAX. Inside tabledist(), switch to unsigned 32 bit arithmetic in order to prevent overflows.
--- net/sched/sch_netem.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
net/sched/sch_netem.c +8 -1
@@ -330,7 +330,7 @@ static s64 tabledist(s64 mu, s32 sigma,/* default uniform distribution */ if (dist ==NULL ) -return ((rnd % (2 * sigma)) + mu) - sigma; +return ((rnd % (2 * (u32 )sigma)) + mu) - sigma; t = dist->table[rnd % dist->size]; x = (sigma %NETEM_DIST_SCALE ) * t;
@@ -787,6 +787,10 @@ static void get_slot(struct netem_sched_ q->slot_config.max_packets =INT_MAX ;if (q->slot_config.max_bytes ==0 ) q->slot_config.max_bytes =INT_MAX ; + +/* capping dist_jitter to the range acceptable by tabledist() */ + q->slot_config.dist_jitter =min_t (__s64,INT_MAX ,abs (q->slot_config.dist_jitter)); + q->slot.packets_left = q->slot_config.max_packets; q->slot.bytes_left = q->slot_config.max_bytes; if (q->slot_config.min_delay | q->slot_config.max_delay |
@@ -1011,6 +1015,9 @@ static int netem_change(struct Qdisc *scif (tb[TCA_NETEM_SLOT ])get_slot (q, tb[TCA_NETEM_SLOT ]); +/* capping jitter to the range acceptable by tabledist() */ + q->jitter =min_t (s64,abs (q->jitter),INT_MAX ); +return ret; get_table_failure:
Greg Kroah-Hartman
maintainerPATCH 13/191 ravb: Fix bit fields checking in ravb_hwtstamp_get()
In reply to Greg Kroah-HartmanFrom: Andrew Gabbasov <[email protected]>
[ Upstream commit 68b9f0865b1ef545da180c57d54b82c94cb464a4 ]
In the function ravb_hwtstamp_get() in ravb_main.c with the existing values for RAVB_RXTSTAMP_TYPE_V2_L2_EVENT (0x2) and RAVB_RXTSTAMP_TYPE_ALL (0x6)
if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_V2_L2_EVENT)
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; else if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_ALL) config.rx_filter = HWTSTAMP_FILTER_ALL;
if the test on RAVB_RXTSTAMP_TYPE_ALL should be true, it will never be reached.
This issue can be verified with 'hwtstamp_config' testing program (tools/testing/selftests/net/hwtstamp_config.c). Setting filter type to ALL and subsequent retrieving it gives incorrect value:
$ hwtstamp_config eth0 OFF ALL flags = 0 tx_type = OFF rx_filter = ALL $ hwtstamp_config eth0 flags = 0 tx_type = OFF rx_filter = PTP_V2_L2_EVENT
Correct this by converting if-else's to switch.
--- drivers/net/ethernet/renesas/ravb_main.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/ethernet/renesas/ravb_main.c +7 -3
@@ -1732,12 +1732,16 @@ static int ravb_hwtstamp_get(struct net_ config.flags =0 ; config.tx_type = priv->tstamp_tx_ctrl ?HWTSTAMP_TX_ON :HWTSTAMP_TX_OFF ; -if (priv->tstamp_rx_ctrl &RAVB_RXTSTAMP_TYPE_V2_L2_EVENT ) +switch (priv->tstamp_rx_ctrl &RAVB_RXTSTAMP_TYPE ) { +case RAVB_RXTSTAMP_TYPE_V2_L2_EVENT : config.rx_filter =HWTSTAMP_FILTER_PTP_V2_L2_EVENT ; -else if (priv->tstamp_rx_ctrl &RAVB_RXTSTAMP_TYPE_ALL ) +break ; +case RAVB_RXTSTAMP_TYPE_ALL : config.rx_filter =HWTSTAMP_FILTER_ALL ; -else +break ; +default : config.rx_filter =HWTSTAMP_FILTER_NONE ; + }return copy_to_user (req->ifr_data, &config,sizeof (config)) ? -EFAULT :0 ;
Greg Kroah-Hartman
maintainerPATCH 14/191 tcp: Prevent low rmem stalls with SO_RCVLOWAT.
In reply to Greg Kroah-HartmanFrom: Arjun Roy <[email protected]>
[ Upstream commit 435ccfa894e35e3d4a1799e6ac030e48a7b69ef5 ]
With SO_RCVLOWAT, under memory pressure,
it is possible to enter a state where:
1. We have not received enough bytes to satisfy SO_RCVLOWAT.
2. We have not entered buffer pressure (see tcp_rmem_pressure()).
3. But, we do not have enough buffer space to accept more packets.
In this case, we advertise 0 rwnd (due to #3) but the application does not drain the receive queue (no wakeup because of #1 and #2) so the flow stalls.
Modify the heuristic for SO_RCVLOWAT so that, if we are advertising rwnd<=rcv_mss, force a wakeup to prevent a stall.
Without this patch, setting tcp_rmem to 6143 and disabling TCP autotune causes a stalled flow. With this patch, no stall occurs. This is with RPC-style traffic with large messages.
--- net/ipv4/tcp.c | 2 ++ net/ipv4/tcp_input.c | 3 ++- 2 files changed, 4 insertions(+), 1 deletion(-)
net/ipv4/tcp.c +2 -0
@@ -495,6 +495,8 @@ static inline bool tcp_stream_is_readablreturn true;if (tcp_rmem_pressure (sk))return true; +if (tcp_receive_window (tp) <=inet_csk (sk)->icsk_ack.rcv_mss) +return true; }if (sk->sk_prot->stream_memory_read)return sk->sk_prot->stream_memory_read (sk);
net/ipv4/tcp_input.c +2 -1
@@ -4704,7 +4704,8 @@ void tcp_data_ready(struct sock *sk)int avail = tp->rcv_nxt - tp->copied_seq;if (avail < sk->sk_rcvlowat && !tcp_rmem_pressure (sk) && - !sock_flag (sk,SOCK_DONE )) + !sock_flag (sk,SOCK_DONE ) && +tcp_receive_window (tp) >inet_csk (sk)->icsk_ack.rcv_mss)return ; sk->sk_data_ready (sk);
Greg Kroah-Hartman
maintainerPATCH 15/191 tipc: fix memory leak caused by tipc_buf_append()
In reply to Greg Kroah-HartmanFrom: Tung Nguyen <[email protected]>
[ Upstream commit ceb1eb2fb609c88363e06618b8d4bbf7815a4e03 ]
Commit ed42989eab57 ("tipc: fix the skb_unshare() in tipc_buf_append()") replaced skb_unshare() with skb_copy() to not reduce the data reference counter of the original skb intentionally. This is not the correct way to handle the cloned skb because it causes memory leak in 2 following cases:
1/ Sending multicast messages via broadcast link
The original skb list is cloned to the local skb list for local destination. After that, the data reference counter of each skb in the original list has the value of 2. This causes each skb not to be freed after receiving ACK:
tipc_link_advance_transmq()
{
...
/* release skb */
__skb_unlink(skb, &l->transmq);
kfree_skb(skb); <-- memory exists after being freed
}
2/ Sending multicast messages via replicast link
Similar to the above case, each skb cannot be freed after purging the skb list:
tipc_mcast_xmit()
{
...
__skb_queue_purge(pkts); <-- memory exists after being freed
}
This commit fixes this issue by using skb_unshare() instead. Besides, to avoid use-after-free error reported by KASAN, the pointer to the fragment is set to NULL before calling skb_unshare() to make sure that the original skb is not freed after freeing the fragment 2 times in case skb_unshare() returns NULL.
--- net/tipc/msg.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
net/tipc/msg.c +2 -3
@@ -140,12 +140,11 @@ int tipc_buf_append(struct sk_buff **heaif (fragid ==FIRST_FRAGMENT ) {if (unlikely (head)) goto err; -if (skb_cloned (frag)) - frag =skb_copy (frag,GFP_ATOMIC ); + *buf =NULL ; + frag =skb_unshare (frag,GFP_ATOMIC );if (unlikely (!frag)) goto err; head = *headbuf = frag; - *buf =NULL ;TIPC_SKB_CB (head)->tail =NULL ;if (skb_is_nonlinear (head)) { skb_walk_frags(head, tail) {
Greg Kroah-Hartman
maintainerPATCH 16/191 r8169: fix issue with forced threading in combination with shared interrupts
In reply to Greg Kroah-HartmanFrom: Heiner Kallweit <[email protected]>
[ Upstream commit 2734a24e6e5d18522fbf599135c59b82ec9b2c9e ]
As reported by Serge flag IRQF_NO_THREAD causes an error if the interrupt is actually shared and the other driver(s) don't have this flag set. This situation can occur if a PCI(e) legacy interrupt is used in combination with forced threading.
There's no good way to deal with this properly, therefore we have to remove flag IRQF_NO_THREAD. For fixing the original forced threading issue switch to napi_schedule().
--- drivers/net/ethernet/realtek/r8169.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/realtek/r8169.c +2 -2
@@ -6630,7 +6630,7 @@ static irqreturn_t rtl8169_interrupt(intreturn IRQ_NONE ;rtl_irq_disable (tp); -napi_schedule_irqoff (&tp->napi); +napi_schedule (&tp->napi);return IRQ_HANDLED ; }
@@ -6886,7 +6886,7 @@ static int rtl_open(struct net_device *drtl_request_firmware (tp); retval =request_irq (pci_irq_vector (pdev,0 ), rtl8169_interrupt, -IRQF_NO_THREAD |IRQF_SHARED , dev->name, tp); +IRQF_SHARED , dev->name, tp);if (retval <0 ) goto err_release_fw_2;
Greg Kroah-Hartman
maintainerPATCH 17/191 cxgb4: set up filter action after rewrites
In reply to Greg Kroah-HartmanFrom: Raju Rangoju <[email protected]>
[ Upstream commit 937d8420588421eaa5c7aa5c79b26b42abb288ef ]
The current code sets up the filter action field before rewrites are set up. When the action 'switch' is used with rewrites, this may result in initial few packets that get switched out don't have rewrites applied on them.
So, make sure filter action is set up along with rewrites or only after everything else is set up for rewrites.
--- drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 56 ++++++++++------------ drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h | 4 + 2 files changed, 31 insertions(+), 29 deletions(-)
drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +27 -29
@@ -145,13 +145,13 @@ static int configure_filter_smac(structint err;/* do a set-tcb for smac-sel and CWR bit.. */ - err =set_tcb_tflag (adap, f, f->tid,TF_CCTRL_CWR_S ,1 ,1 ); -if (err) - goto smac_err; - err =set_tcb_field (adap, f, f->tid,TCB_SMAC_SEL_W ,TCB_SMAC_SEL_V (TCB_SMAC_SEL_M ),TCB_SMAC_SEL_V (f->smt->idx),1 ); +if (err) + goto smac_err; + + err =set_tcb_tflag (adap, f, f->tid,TF_CCTRL_CWR_S ,1 ,1 );if (!err)return 0 ;
@@ -608,6 +608,7 @@ int set_filter_wr(struct adapter *adapteFW_FILTER_WR_DIRSTEERHASH_V (f->fs.dirsteerhash) |FW_FILTER_WR_LPBK_V (f->fs.action ==FILTER_SWITCH ) |FW_FILTER_WR_DMAC_V (f->fs.newdmac) | +FW_FILTER_WR_SMAC_V (f->fs.newsmac) |FW_FILTER_WR_INSVLAN_V (f->fs.newvlan ==VLAN_INSERT || f->fs.newvlan ==VLAN_REWRITE ) | FW_FILTER_WR_RMVLAN_V(f->fs.newvlan == VLAN_REMOVE ||
@@ -625,7 +626,7 @@ int set_filter_wr(struct adapter *adapteFW_FILTER_WR_OVLAN_VLD_V (f->fs.val.ovlan_vld) |FW_FILTER_WR_IVLAN_VLDM_V (f->fs.mask.ivlan_vld) | FW_FILTER_WR_OVLAN_VLDM_V(f->fs.mask.ovlan_vld)); - fwr->smac_sel =0 ; + fwr->smac_sel = f->smt->idx; fwr->rx_chan_rx_rpl_iq =htons (FW_FILTER_WR_RX_CHAN_V (0 ) |FW_FILTER_WR_RX_RPL_IQ_V (adapter->sge.fw_evtq.abs_id));
@@ -1019,11 +1020,8 @@ static void mk_act_open_req6(struct filtTX_QUEUE_V (f->fs.nat_mode) |T5_OPT_2_VALID_F |RX_CHANNEL_F | -CONG_CNTRL_V ((f->fs.action ==FILTER_DROP ) | - (f->fs.dirsteer <<1 )) |PACE_V ((f->fs.maskhash) | - ((f->fs.dirsteerhash) <<1 )) | - CCTRL_ECN_V(f->fs.action == FILTER_SWITCH)); + ((f->fs.dirsteerhash) << 1))); } static void mk_act_open_req(struct filter_entry *f, struct sk_buff *skb,
@@ -1059,11 +1057,8 @@ static void mk_act_open_req(struct filteTX_QUEUE_V (f->fs.nat_mode) |T5_OPT_2_VALID_F |RX_CHANNEL_F | -CONG_CNTRL_V ((f->fs.action ==FILTER_DROP ) | - (f->fs.dirsteer <<1 )) |PACE_V ((f->fs.maskhash) | - ((f->fs.dirsteerhash) <<1 )) | - CCTRL_ECN_V(f->fs.action == FILTER_SWITCH)); + ((f->fs.dirsteerhash) << 1))); } static int cxgb4_set_hash_filter(struct net_device *dev,
@@ -1722,6 +1717,20 @@ void hash_filter_rpl(struct adapter *ada }return ; } +switch (f->fs.action) { +case FILTER_PASS : +if (f->fs.dirsteer) +set_tcb_tflag (adap, f, tid, +TF_DIRECT_STEER_S ,1 ,1 ); +break ; +case FILTER_DROP : +set_tcb_tflag (adap, f, tid,TF_DROP_S ,1 ,1 ); +break ; +case FILTER_SWITCH : +set_tcb_tflag (adap, f, tid,TF_LPBK_S ,1 ,1 ); +break ; + } +break ;default :
@@ -1781,22 +1790,11 @@ void filter_rpl(struct adapter *adap, coif (ctx) ctx->result =0 ; }else if (ret ==FW_FILTER_WR_FLT_ADDED ) { -int err =0 ; - -if (f->fs.newsmac) - err =configure_filter_smac (adap, f); - -if (!err) { - f->pending =0 ;/* async setup completed */ - f->valid =1 ; -if (ctx) { - ctx->result =0 ; - ctx->tid = idx; - } - }else { -clear_filter (adap, f); -if (ctx) - ctx->result = err; + f->pending =0 ;/* async setup completed */ + f->valid =1 ; +if (ctx) { + ctx->result =0 ; + ctx->tid = idx; } }else {/* Something went wrong. Issue a warning about the
drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h +4 -0
@@ -42,6 +42,10 @@#define TCB_T_FLAGS_W 1 +#define TF_DROP_S 22 +#define TF_DIRECT_STEER_S 23 +#define TF_LPBK_S 59 +#define TF_CCTRL_ECE_S 60#define TF_CCTRL_CWR_S 61#define TF_CCTRL_RFR_S 62
Greg Kroah-Hartman
maintainerPATCH 18/191 arch/x86/amd/ibs: Fix re-arming IBS Fetch
In reply to Greg Kroah-HartmanFrom: Kim Phillips <[email protected]>
commit 221bfce5ebbdf72ff08b3bf2510ae81058ee568b upstream.
Stephane Eranian found a bug in that IBS' current Fetch counter was not being reset when the driver would write the new value to clear it along with the enable bit set, and found that adding an MSR write that would first disable IBS Fetch would make IBS Fetch reset its current count.
Indeed, the PPR for AMD Family 17h Model 31h B0 55803 Rev 0.54 - Sep 12, 2019 states "The periodic fetch counter is set to IbsFetchCnt [...] when IbsFetchEn is changed from 0 to 1."
Explicitly set IbsFetchEn to 0 and then to 1 when re-enabling IBS Fetch, so the driver properly resets the internal counter to 0 and IBS Fetch starts counting again.
A family 15h machine tested does not have this problem, and the extra wrmsr is also not needed on Family 19h, so only do the extra wrmsr on families 16h through 18h.
--- arch/x86/events/amd/ibs.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
arch/x86/events/amd/ibs.c +14 -1
@@ -89,6 +89,7 @@ struct perf_ibs {u64 max_period;unsigned long offset_mask[1 ];int offset_max; + unsigned int fetch_count_reset_broken : 1;struct cpu_perf_ibs __percpu *pcpu;struct attribute **format_attrs;
@@ -375,7 +376,12 @@ perf_ibs_event_update(struct perf_ibs *pstatic inline void perf_ibs_enable_event (struct perf_ibs *perf_ibs,struct hw_perf_event *hwc,u64 config) { -wrmsrl (hwc->config_base, hwc->config | config | perf_ibs->enable_mask); +u64 tmp = hwc->config | config; + +if (perf_ibs->fetch_count_reset_broken) +wrmsrl (hwc->config_base, tmp & ~perf_ibs->enable_mask); + +wrmsrl (hwc->config_base, tmp | perf_ibs->enable_mask); }/*
@@ -744,6 +750,13 @@ static __init void perf_event_ibs_init(v {struct attribute **attr = ibs_op_format_attrs; +/* +* Some chips fail to reset the fetch count when it is written; instead +* they need a 0-1 transition of IbsFetchEn. +*/ +if (boot_cpu_data.x86 >=0x16 && boot_cpu_data.x86 <=0x18 ) + perf_ibs_fetch.fetch_count_reset_broken =1 ; +perf_ibs_pmu_init (&perf_ibs_fetch,"ibs_fetch" );if (ibs_caps &IBS_CAPS_OPCNT ) {
Greg Kroah-Hartman
maintainerPATCH 19/191 x86/xen: disable Firmware First mode for correctable memory errors
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit d759af38572f97321112a0852353613d18126038 upstream.
When running as Xen dom0 the kernel isn't responsible for selecting the error handling mode, this should be handled by the hypervisor.
So disable setting FF mode when running as Xen pv guest. Not doing so might result in boot splats like:
[ 7.509696] HEST: Enabling Firmware First mode for corrected errors. [ 7.510382] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 2. [ 7.510383] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 3. [ 7.510384] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 4. [ 7.510384] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 5. [ 7.510385] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 6. [ 7.510386] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 7. [ 7.510386] mce: [Firmware Bug]: Ignoring request to disable invalid MCA bank 8.
Reason is that the HEST ACPI table contains the real number of MCA banks, while the hypervisor is emulating only 2 banks for guests.
--- arch/x86/xen/enlighten_pv.c | 9 +++++++++ 1 file changed, 9 insertions(+)
arch/x86/xen/enlighten_pv.c +9 -0
@@ -1383,6 +1383,15 @@ asmlinkage __visible void __init xen_sta x86_init.mpparse.get_smp_config = x86_init_uint_noop;xen_boot_params_init_edd (); + +#ifdef CONFIG_ACPI +/* +* Disable selecting "Firmware First mode" for correctable +* memory errors, as this is the duty of the hypervisor to +* decide. +*/ + acpi_disable_cmcff =1 ; +#endif }if (!boot_params.screen_info.orig_video_isVGA)
Greg Kroah-Hartman
maintainerPATCH 20/191 fuse: fix page dereference after free
In reply to Greg Kroah-HartmanFrom: Miklos Szeredi <[email protected]>
commit d78092e4937de9ce55edcb4ee4c5e3c707be0190 upstream.
After unlock_request() pages from the ap->pages[] array may be put (e.g. by aborting the connection) and the pages can be freed.
Prevent use after free by grabbing a reference to the page before calling unlock_request().
The original patch was created by Pradeep P V K.
--- fs/fuse/dev.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-)
fs/fuse/dev.c +18 -10
@@ -853,15 +853,16 @@ static int fuse_try_move_page(struct fusstruct page *newpage;struct pipe_buffer *buf = cs->pipebufs; +get_page (oldpage); err =unlock_request (cs->req);if (err) -return err; + goto out_put_old;fuse_copy_finish (cs); err =pipe_buf_confirm (cs->pipe, buf);if (err) -return err; + goto out_put_old;BUG_ON (!cs->nr_segs); cs->currbuf = buf;
@@ -901,7 +902,7 @@ static int fuse_try_move_page(struct fus err =replace_page_cache_page (oldpage, newpage,GFP_KERNEL );if (err) {unlock_page (newpage); -return err; + goto out_put_old; }get_page (newpage);
@@ -920,14 +921,19 @@ static int fuse_try_move_page(struct fusif (err) {unlock_page (newpage);put_page (newpage); -return err; + goto out_put_old; }unlock_page (oldpage); +/* Drop ref for ap->pages[] array */ put_page (oldpage); cs->len =0 ; -return 0 ; + err =0 ; +out_put_old: +/* Drop ref obtained in this function */ +put_page (oldpage); +return err; out_fallback_unlock:unlock_page (newpage);
@@ -936,10 +942,10 @@ out_fallback: cs->offset = buf->offset; err =lock_request (cs->req); -if (err) -return err; +if (!err) + err =1 ; -return 1 ; + goto out_put_old; } static int fuse_ref_page(struct fuse_copy_state *cs, struct page *page,
@@ -951,14 +957,16 @@ static int fuse_ref_page(struct fuse_copif (cs->nr_segs == cs->pipe->buffers)return -EIO ; +get_page (page); err =unlock_request (cs->req); -if (err) +if (err) { +put_page (page);return err; + }fuse_copy_finish (cs); buf = cs->pipebufs; -get_page (page); buf->page = page; buf->offset = offset; buf->len = count;
Greg Kroah-Hartman
maintainerPATCH 21/191 bpf: Fix comment for helper bpf_current_task_under_cgroup()
In reply to Greg Kroah-HartmanFrom: Song Liu <[email protected]>
commit 1aef5b4391f0c75c0a1523706a7b0311846ee12f upstream.
This should be "current" not "skb".
--- include/uapi/linux/bpf.h | 4 ++-- tools/include/uapi/linux/bpf.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-)
include/uapi/linux/bpf.h +2 -2
@@ -1193,8 +1193,8 @@ union bpf_attr {* Return * The return value depends on the result of the test, and can be: * -* * 0, if the *skb* task belongs to the cgroup2. -* * 1, if the *skb* task does not belong to the cgroup2. +* * 0, if current task belongs to the cgroup2. +* * 1, if current task does not belong to the cgroup2. * * A negative error code, if an error occurred. * * int bpf_skb_change_tail(struct sk_buff *skb, u32 len, u64 flags)
tools/include/uapi/linux/bpf.h +2 -2
@@ -1191,8 +1191,8 @@ union bpf_attr {* Return * The return value depends on the result of the test, and can be: * -* * 0, if the *skb* task belongs to the cgroup2. -* * 1, if the *skb* task does not belong to the cgroup2. +* * 0, if current task belongs to the cgroup2. +* * 1, if current task does not belong to the cgroup2. * * A negative error code, if an error occurred. * * int bpf_skb_change_tail(struct sk_buff *skb, u32 len, u64 flags)
Greg Kroah-Hartman
maintainerPATCH 22/191 evm: Check size of security.evm before using it
In reply to Greg Kroah-HartmanFrom: Roberto Sassu <[email protected]>
commit 455b6c9112eff8d249e32ba165742085678a80a4 upstream.
This patch checks the size for the EVM_IMA_XATTR_DIGSIG and EVM_XATTR_PORTABLE_DIGSIG types to ensure that the algorithm is read from the buffer returned by vfs_getxattr_alloc().
--- security/integrity/evm/evm_main.c | 6 ++++++ 1 file changed, 6 insertions(+)
security/integrity/evm/evm_main.c +6 -0
@@ -186,6 +186,12 @@ static enum integrity_status evm_verify_break ;case EVM_IMA_XATTR_DIGSIG :case EVM_XATTR_PORTABLE_DIGSIG : +/* accept xattr with non-empty signature field */ +if (xattr_len <=sizeof (struct signature_v2_hdr )) { + evm_status =INTEGRITY_FAIL ; + goto out; + } + hdr = (struct signature_v2_hdr *)xattr_data; digest.hdr.algo = hdr->hash_algo; rc = evm_calc_hash(dentry, xattr_name, xattr_value,
Greg Kroah-Hartman
maintainerPATCH 23/191 p54: avoid accessing the data mapped to streaming DMA
In reply to Greg Kroah-HartmanFrom: Jia-Ju Bai <[email protected]>
commit 478762855b5ae9f68fa6ead1edf7abada70fcd5f upstream.
In p54p_tx(), skb->data is mapped to streaming DMA on line 337:
mapping = pci_map_single(..., skb->data, ...);
Then skb->data is accessed on line 349:
desc->device_addr = ((struct p54_hdr *)skb->data)->req_id;
This access may cause data inconsistency between CPU cache and hardware.
To fix this problem, ((struct p54_hdr *)skb->data)->req_id is stored in a local variable before DMA mapping, and then the driver accesses this local variable instead of skb->data.
--- drivers/net/wireless/intersil/p54/p54pci.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/wireless/intersil/p54/p54pci.c +3 -1
@@ -332,10 +332,12 @@ static void p54p_tx(struct ieee80211_hwstruct p54p_desc *desc;dma_addr_t mapping;u32 idx, i; +__le32 device_addr;spin_lock_irqsave (&priv->lock, flags); idx =le32_to_cpu (ring_control->host_idx[1 ]); i = idx %ARRAY_SIZE (ring_control->tx_data); + device_addr = ((struct p54_hdr *)skb->data)->req_id; mapping =pci_map_single (priv->pdev, skb->data, skb->len,PCI_DMA_TODEVICE );
@@ -349,7 +351,7 @@ static void p54p_tx(struct ieee80211_hw desc = &ring_control->tx_data[i]; desc->host_addr =cpu_to_le32 (mapping); - desc->device_addr = ((struct p54_hdr *)skb->data)->req_id; + desc->device_addr = device_addr; desc->len =cpu_to_le16 (skb->len); desc->flags =0 ;
Greg Kroah-Hartman
maintainerPATCH 24/191 cxl: Rework error message for incompatible slots
In reply to Greg Kroah-HartmanFrom: Frederic Barrat <[email protected]>
commit 40ac790d99c6dd16b367d5c2339e446a5f1b0593 upstream.
Improve the error message shown if a capi adapter is plugged on a capi-incompatible slot directly under the PHB (no intermediate switch).
--- drivers/misc/cxl/pci.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/misc/cxl/pci.c +2 -2
@@ -397,8 +397,8 @@ int cxl_calc_capp_routing(struct pci_dev *capp_unit_id =get_capp_unit_id (np, *phb_index);of_node_put (np);if (!*capp_unit_id) { -pr_err ("cxl: invalid capp unit id (phb_index: %d)\n" , - *phb_index); +pr_err ("cxl: No capp unit found for PHB[%lld,%d]. Make sure the adapter is on a capi-compatible slot\n" , + *chipid, *phb_index);return -ENODEV ; }
Greg Kroah-Hartman
maintainerPATCH 25/191 RDMA/addr: Fix race with netevent_callback()/rdma_addr_cancel()
In reply to Greg Kroah-HartmanFrom: Jason Gunthorpe <[email protected]>
commit 2ee9bf346fbfd1dad0933b9eb3a4c2c0979b633e upstream.
This three thread race can result in the work being run once the callback becomes NULL:
CPU1 CPU2 CPU3
netevent_callback()
process_one_req() rdma_addr_cancel()
[..]
spin_lock_bh()
set_timeout()
spin_unlock_bh()
spin_lock_bh()
list_del_init(&req->list);
spin_unlock_bh()
req->callback = NULL
spin_lock_bh()
if (!list_empty(&req->list))
// Skipped!
// cancel_delayed_work(&req->work);
spin_unlock_bh()
process_one_req() // again
req->callback() // BOOM
cancel_delayed_work_sync()The solution is to always cancel the work once it is completed so any in between set_timeout() does not result in it running again.
--- drivers/infiniband/core/addr.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/infiniband/core/addr.c +5 -6
@@ -571,13 +571,12 @@ static void process_one_req(struct work_ req->callback =NULL ;spin_lock_bh (&lock); +/* +* Although the work will normally have been canceled by the workqueue, +* it can still be requeued as long as it is on the req_list. +*/ +cancel_delayed_work (&req->work);if (!list_empty (&req->list)) { -/* -* Although the work will normally have been canceled by the -* workqueue, it can still be requeued as long as it is on the -* req_list. -*/ -cancel_delayed_work (&req->work);list_del_init (&req->list);kfree (req); }
Greg Kroah-Hartman
maintainerPATCH 26/191 mtd: lpddr: Fix bad logic in print_drs_error
In reply to Greg Kroah-HartmanFrom: Gustavo A. R. Silva <[email protected]>
commit 1c9c02bb22684f6949d2e7ddc0a3ff364fd5a6fc upstream.
Update logic for broken test. Use a more common logging style.
It appears the logic in this function is broken for the consecutive tests of
if (prog_status & 0x3)
...
else if (prog_status & 0x2)
...
else (prog_status & 0x1)
...Likely the first test should be
if ((prog_status & 0x3) == 0x3)
Found by inspection of include files using printk.
--- include/linux/mtd/pfow.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/mtd/pfow.h +1 -1
@@ -128,7 +128,7 @@ static inline void print_drs_error(unsigif (!(dsr &DSR_AVAILABLE ))printk (KERN_NOTICE "DSR.15: (0) Device not Available\n" ); -if (prog_status &0x03 ) +if ((prog_status &0x03 ) ==0x03 )printk (KERN_NOTICE "DSR.9,8: (11) Attempt to program invalid " "half with 41h command\n" );else if (prog_status &0x02 )
Greg Kroah-Hartman
maintainerPATCH 27/191 serial: pl011: Fix lockdep splat when handling magic-sysrq interrupt
In reply to Greg Kroah-HartmanFrom: Peter Zijlstra <[email protected]>
commit 534cf755d9df99e214ddbe26b91cd4d81d2603e2 upstream.
Issuing a magic-sysrq via the PL011 causes the following lockdep splat, which is easily reproducible under QEMU:
| sysrq: Changing Loglevel
| sysrq: Loglevel set to 9
|
| ======================================================
| WARNING: possible circular locking dependency detected
| 5.9.0-rc7 #1 Not tainted
| ------------------------------------------------------
| systemd-journal/138 is trying to acquire lock:
| ffffab133ad950c0 (console_owner){-.-.}-{0:0}, at: console_lock_spinning_enable+0x34/0x70
|
| but task is already holding lock:
| ffff0001fd47b098 (&port_lock_key){-.-.}-{2:2}, at: pl011_int+0x40/0x488
|
| which lock already depends on the new lock.[...]
| Possible unsafe locking scenario: | | CPU0 CPU1 | ---- ---- | lock(&port_lock_key); | lock(console_owner); | lock(&port_lock_key); | lock(console_owner); | | *** DEADLOCK ***
The issue being that CPU0 takes 'port_lock' on the irq path in pl011_int() before taking 'console_owner' on the printk() path, whereas CPU1 takes the two locks in the opposite order on the printk() path due to setting the "console_owner" prior to calling into into the actual console driver.
Fix this in the same way as the msm-serial driver by dropping 'port_lock' before handling the sysrq.
--- drivers/tty/serial/amba-pl011.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/tty/serial/amba-pl011.c +7 -4
@@ -313,8 +313,9 @@ static void pl011_write(unsigned int val*/ static int pl011_fifo_to_tty (struct uart_amba_port *uap) { -u16 status;unsigned int ch, flag, fifotaken; +int sysrq; +u16 status;for (fifotaken =0 ; fifotaken !=256 ; fifotaken++) { status =pl011_read (uap,REG_FR );
@@ -349,10 +350,12 @@ static int pl011_fifo_to_tty(struct uart flag =TTY_FRAME ; } -if (uart_handle_sysrq_char (&uap->port, ch &255 )) -continue ; +spin_unlock (&uap->port.lock); + sysrq =uart_handle_sysrq_char (&uap->port, ch &255 ); +spin_lock (&uap->port.lock); -uart_insert_char (&uap->port, ch,UART011_DR_OE , ch, flag); +if (!sysrq) +uart_insert_char (&uap->port, ch,UART011_DR_OE , ch, flag); }return fifotaken;
Greg Kroah-Hartman
maintainerPATCH 28/191 ata: sata_rcar: Fix DMA boundary mask
In reply to Greg Kroah-HartmanFrom: Geert Uytterhoeven <[email protected]>
commit df9c590986fdb6db9d5636d6cd93bc919c01b451 upstream.
Before commit 9495b7e92f716ab2 ("driver core: platform: Initialize dma_parms for platform devices"), the R-Car SATA device didn't have DMA parameters. Hence the DMA boundary mask supplied by its driver was silently ignored, as __scsi_init_queue() doesn't check the return value of dma_set_seg_boundary(), and the default value of 0xffffffff was used.
Now the device has gained DMA parameters, the driver-supplied value is used, and the following warning is printed on Salvator-XS:
DMA-API: sata_rcar ee300000.sata: mapping sg segment across boundary [start=0x00000000ffffe000] [end=0x00000000ffffefff] [boundary=0x000000001ffffffe]
WARNING: CPU: 5 PID: 38 at kernel/dma/debug.c:1233 debug_dma_map_sg+0x298/0x300(the range of start/end values depend on whether IOMMU support is
enabled or not)
The issue here is that SATA_RCAR_DMA_BOUNDARY doesn't have bit 0 set, so any typical end value, which is odd, will trigger the check.
Fix this by increasing the DMA boundary value by 1.
This also fixes the following WRITE DMA EXT timeout issue:
# dd if=/dev/urandom of=/mnt/de1/file1-1024M bs=1M count=1024
ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
ata1.00: failed command: WRITE DMA EXT
ata1.00: cmd 35/00:00:00:e6:0c/00:0a:00:00:00/e0 tag 0 dma 1310720 out
res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
ata1.00: status: { DRDY }as seen by Shimoda-san since commit 429120f3df2dba2b ("block: fix splitting segments on boundary masks").
--- drivers/ata/sata_rcar.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/sata_rcar.c +1 -1
@@ -124,7 +124,7 @@/* Descriptor table word 0 bit (when DTA32M = 1) */ #define SATA_RCAR_DTEND BIT(0) -#define SATA_RCAR_DMA_BOUNDARY 0x1FFFFFFEUL +#define SATA_RCAR_DMA_BOUNDARY 0x1FFFFFFFUL/* Gen2 Physical Layer Control Registers */ #define RCAR_GEN2_PHY_CTL1_REG 0x1704
Greg Kroah-Hartman
maintainerPATCH 29/191 fscrypt: return -EXDEV for incompatible rename or link into encrypted dir
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit f5e55e777cc93eae1416f0fa4908e8846b6d7825 upstream.
Currently, trying to rename or link a regular file, directory, or symlink into an encrypted directory fails with EPERM when the source file is unencrypted or is encrypted with a different encryption policy, and is on the same mountpoint. It is correct for the operation to fail, but the choice of EPERM breaks tools like 'mv' that know to copy rather than rename if they see EXDEV, but don't know what to do with EPERM.
Our original motivation for EPERM was to encourage users to securely handle their data. Encrypting files by "moving" them into an encrypted directory can be insecure because the unencrypted data may remain in free space on disk, where it can later be recovered by an attacker. It's much better to encrypt the data from the start, or at least try to securely delete the source data e.g. using the 'shred' program.
However, the current behavior hasn't been effective at achieving its goal because users tend to be confused, hack around it, and complain; see e.g. https://github.com/google/fscrypt/issues/76. And in some cases it's actually inconsistent or unnecessary. For example, 'mv'-ing files between differently encrypted directories doesn't work even in cases where it can be secure, such as when in userspace the same passphrase protects both directories. Yet, you *can* already 'mv' unencrypted files into an encrypted directory if the source files are on a different mountpoint, even though doing so is often insecure.
There are probably better ways to teach users to securely handle their files. For example, the 'fscrypt' userspace tool could provide a command that migrates unencrypted files into an encrypted directory, acting like 'shred' on the source files and providing appropriate warnings depending on the type of the source filesystem and disk.
Receiving errors on unimportant files might also force some users to disable encryption, thus making the behavior counterproductive. It's desirable to make encryption as unobtrusive as possible.
Therefore, change the error code from EPERM to EXDEV so that tools looking for EXDEV will fall back to a copy.
This, of course, doesn't prevent users from still doing the right things to securely manage their files. Note that this also matches the behavior when a file is renamed between two project quota hierarchies; so there's precedent for using EXDEV for things other than mountpoints.
xfstests generic/398 will require an update with this change.
[Rewritten from an earlier patch series by Michael Halcrow.]
--- Documentation/filesystems/fscrypt.rst | 12 ++++++++++-- fs/crypto/hooks.c | 6 +++--- fs/crypto/policy.c | 3 +-- include/linux/fscrypt.h | 4 ++-- 4 files changed, 16 insertions(+), 9 deletions(-)
Documentation/filesystems/fscrypt.rst +10 -2
@@ -426,10 +426,18 @@ astute users may notice some differences - Unencrypted files, or files encrypted with a different encryption policy (i.e. different key, modes, or flags), cannot be renamed or linked into an encrypted directory; see `Encryption policy - enforcement`_. Attempts to do so will fail with EPERM. However, + enforcement`_. Attempts to do so will fail with EXDEV. However, encrypted files can be renamed within an encrypted directory, or into an unencrypted directory. + Note: "moving" an unencrypted file into an encrypted directory, e.g. + with the `mv` program, is implemented in userspace by a copy + followed by a delete. Be aware that the original unencrypted data + may remain recoverable from free space on the disk; prefer to keep + all files encrypted from the very beginning. The `shred` program + may be used to overwrite the source files but isn't guaranteed to be + effective on all filesystems and storage devices. + - Direct I/O is not supported on encrypted files. Attempts to use direct I/O on such files will fall back to buffered I/O.
@@ -516,7 +524,7 @@ not be encrypted. Except for those special files, it is forbidden to have unencrypted files, or files encrypted with a different encryption policy, in an encrypted directory tree. Attempts to link or rename such a file into -an encrypted directory will fail with EPERM. This is also enforced +an encrypted directory will fail with EXDEV. This is also enforced during ->lookup() to provide limited protection against offline attacks that try to disable or downgrade encryption in known locations where applications may later write sensitive data. It is recommended
fs/crypto/hooks.c +3 -3
@@ -58,7 +58,7 @@ int __fscrypt_prepare_link(struct inodereturn err;if (!fscrypt_has_permitted_context (dir, inode)) -return -EPERM ; +return -EXDEV ;return 0 ; }
@@ -82,13 +82,13 @@ int __fscrypt_prepare_rename(struct inodif (IS_ENCRYPTED (new_dir) && !fscrypt_has_permitted_context (new_dir,d_inode (old_dentry))) -return -EPERM ; +return -EXDEV ;if ((flags &RENAME_EXCHANGE ) &&IS_ENCRYPTED (old_dir) && !fscrypt_has_permitted_context (old_dir,d_inode (new_dentry))) -return -EPERM ; +return -EXDEV ; }return 0 ; }
fs/crypto/policy.c +1 -2
@@ -153,8 +153,7 @@ EXPORT_SYMBOL(fscrypt_ioctl_get_policy);* malicious offline violations of this constraint, while the link and rename * checks are needed to prevent online violations of this constraint. * -* Return: 1 if permitted, 0 if forbidden. If forbidden, the caller must fail -* the filesystem operation with EPERM. +* Return: 1 if permitted, 0 if forbidden. */ int fscrypt_has_permitted_context (struct inode *parent,struct inode *child) {
include/linux/fscrypt.h +2 -2
@@ -89,7 +89,7 @@ static inline int fscrypt_require_key(st* in an encrypted directory tree use the same encryption policy. * * Return: 0 on success, -ENOKEY if the directory's encryption key is missing, -* -EPERM if the link would result in an inconsistent encryption policy, or +* -EXDEV if the link would result in an inconsistent encryption policy, or * another -errno code. */ static inline int fscrypt_prepare_link(struct dentry *old_dentry,
@@ -119,7 +119,7 @@ static inline int fscrypt_prepare_link(s* We also verify that the rename will not violate the constraint that all files * in an encrypted directory tree use the same encryption policy. * -* Return: 0 on success, -ENOKEY if an encryption key is missing, -EPERM if the +* Return: 0 on success, -ENOKEY if an encryption key is missing, -EXDEV if the * rename would cause inconsistent encryption policies, or another -errno code. */ static inline int fscrypt_prepare_rename(struct inode *old_dir,
Greg Kroah-Hartman
maintainerPATCH 30/191 fscrypt: clean up and improve dentry revalidation
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit 6cc248684d3d23bbd073ae2fa73d3416c0558909 upstream.
Make various improvements to fscrypt dentry revalidation:
- Don't try to handle the case where the per-directory key is removed, as this can't happen without the inode (and dentries) being evicted.
- Flag ciphertext dentries rather than plaintext dentries, since it's ciphertext dentries that need the special handling.
- Avoid doing unnecessary work for non-ciphertext dentries.
- When revalidating ciphertext dentries, try to set up the directory's i_crypt_info to make sure the key is really still absent, rather than invalidating all negative dentries as the previous code did. An old comment suggested we can't do this for locking reasons, but AFAICT this comment was outdated and it actually works fine.
--- fs/crypto/crypto.c | 58 ++++++++++++++++++++++++------------------------ fs/crypto/hooks.c | 4 +-- include/linux/dcache.h | 2 - include/linux/fscrypt.h | 6 +--- 4 files changed, 35 insertions(+), 35 deletions(-)
fs/crypto/crypto.c +30 -28
@@ -314,45 +314,47 @@ int fscrypt_decrypt_page(const struct inEXPORT_SYMBOL (fscrypt_decrypt_page);/* -* Validate dentries for encrypted directories to make sure we aren't -* potentially caching stale data after a key has been added or -* removed. +* Validate dentries in encrypted directories to make sure we aren't potentially +* caching stale dentries after a key has been added. */ static int fscrypt_d_revalidate (struct dentry *dentry,unsigned int flags) {struct dentry *dir; -int dir_has_key, cached_with_key; +int err; +int valid; + +/* +* Plaintext names are always valid, since fscrypt doesn't support +* reverting to ciphertext names without evicting the directory's inode +* -- which implies eviction of the dentries in the directory. +*/ +if (!(dentry->d_flags &DCACHE_ENCRYPTED_NAME )) +return 1 ; + +/* +* Ciphertext name; valid if the directory's key is still unavailable. +* +* Although fscrypt forbids rename() on ciphertext names, we still must +* use dget_parent() here rather than use ->d_parent directly. That's +* because a corrupted fs image may contain directory hard links, which +* the VFS handles by moving the directory's dentry tree in the dcache +* each time ->lookup() finds the directory and it already has a dentry +* elsewhere. Thus ->d_parent can be changing, and we must safely grab +* a reference to some ->d_parent to prevent it from being freed. +*/ if (flags &LOOKUP_RCU )return -ECHILD ; dir =dget_parent (dentry); -if (!IS_ENCRYPTED (d_inode (dir))) { -dput (dir); -return 0 ; - } - -spin_lock (&dentry->d_lock); - cached_with_key = dentry->d_flags &DCACHE_ENCRYPTED_WITH_KEY ; -spin_unlock (&dentry->d_lock); - dir_has_key = (d_inode (dir)->i_crypt_info !=NULL ); + err =fscrypt_get_encryption_info (d_inode (dir)); + valid = !fscrypt_has_encryption_key (d_inode (dir));dput (dir); -/* -* If the dentry was cached without the key, and it is a -* negative dentry, it might be a valid name. We can't check -* if the key has since been made available due to locking -* reasons, so we fail the validation so ext4_lookup() can do -* this check. -* -* We also fail the validation if the dentry was created with -* the key present, but we no longer have the key, or vice versa. -*/ -if ((!cached_with_key &&d_is_negative (dentry)) || - (!cached_with_key && dir_has_key) || - (cached_with_key && !dir_has_key)) -return 0 ; -return 1 ; +if (err <0 ) +return err; + +return valid; }const struct dentry_operations fscrypt_d_ops = {
fs/crypto/hooks.c +2 -2
@@ -101,9 +101,9 @@ int __fscrypt_prepare_lookup(struct inodif (err)return err; -if (fscrypt_has_encryption_key (dir)) { +if (!fscrypt_has_encryption_key (dir)) {spin_lock (&dentry->d_lock); - dentry->d_flags |=DCACHE_ENCRYPTED_WITH_KEY ; + dentry->d_flags |=DCACHE_ENCRYPTED_NAME ;spin_unlock (&dentry->d_lock); }
include/linux/dcache.h +1 -1
@@ -210,7 +210,7 @@ struct dentry_operations {#define DCACHE_MAY_FREE 0x00800000#define DCACHE_FALLTHRU 0x01000000/* Fall through to lower layer */ -#define DCACHE_ENCRYPTED_WITH_KEY 0x02000000/* dir is encrypted with a valid key */ +#define DCACHE_ENCRYPTED_NAME 0x02000000/* Encrypted name (dir key was unavailable) */ #define DCACHE_OP_REAL 0x04000000#define DCACHE_PAR_LOOKUP 0x10000000/* being looked up (with parent locked shared) */
include/linux/fscrypt.h +2 -4
@@ -145,10 +145,8 @@ static inline int fscrypt_prepare_rename* filenames are presented in encrypted form. Therefore, we'll try to set up * the directory's encryption key, but even without it the lookup can continue. * -* To allow invalidating stale dentries if the directory's encryption key is -* added later, we also install a custom ->d_revalidate() method and use the -* DCACHE_ENCRYPTED_WITH_KEY flag to indicate whether a given dentry is a -* plaintext name (flag set) or a ciphertext name (flag cleared). +* This also installs a custom ->d_revalidate() method which will invalidate the +* dentry if it was created without the key and the key is later added. * * Return: 0 on success, -errno if a problem occurred while setting up the * encryption key
Greg Kroah-Hartman
maintainerPATCH 31/191 fscrypt: fix race allowing rename() and link() of ciphertext dentries
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit 968dd6d0c6d6b6a989c6ddb9e2584a031b83e7b5 upstream.
Close some race conditions where fscrypt allowed rename() and link() on ciphertext dentries that had been looked up just prior to the key being concurrently added. It's better to return -ENOKEY in this case.
This avoids doing the nonsensical thing of encrypting the names a second time when searching for the actual on-disk dir entries. It also guarantees that DCACHE_ENCRYPTED_NAME dentries are never rename()d, so the dcache won't have support all possible combinations of moving DCACHE_ENCRYPTED_NAME around during __d_move().
--- fs/crypto/hooks.c | 12 +++++++++++- include/linux/fscrypt.h | 2 +- include/linux/fscrypt_notsupp.h | 4 ++-- include/linux/fscrypt_supp.h | 3 ++- 4 files changed, 16 insertions(+), 5 deletions(-)
fs/crypto/hooks.c +11 -1
@@ -49,7 +49,8 @@ int fscrypt_file_open(struct inode *inod }EXPORT_SYMBOL_GPL (fscrypt_file_open); -int __fscrypt_prepare_link (struct inode *inode,struct inode *dir) +int __fscrypt_prepare_link (struct inode *inode,struct inode *dir, +struct dentry *dentry) {int err;
@@ -57,6 +58,10 @@ int __fscrypt_prepare_link(struct inodeif (err)return err; +/* ... in case we looked up ciphertext name before key was added */ +if (dentry->d_flags &DCACHE_ENCRYPTED_NAME ) +return -ENOKEY ; +if (!fscrypt_has_permitted_context (dir, inode))return -EXDEV ;
@@ -78,6 +83,11 @@ int __fscrypt_prepare_rename(struct inodif (err)return err; +/* ... in case we looked up ciphertext name(s) before key was added */ +if ((old_dentry->d_flags | new_dentry->d_flags) & +DCACHE_ENCRYPTED_NAME ) +return -ENOKEY ; + if (old_dir != new_dir) { if (IS_ENCRYPTED(new_dir) && !fscrypt_has_permitted_context(new_dir,
include/linux/fscrypt.h +1 -1
@@ -97,7 +97,7 @@ static inline int fscrypt_prepare_link(s struct dentry *dentry) {if (IS_ENCRYPTED (dir)) -return __fscrypt_prepare_link (d_inode (old_dentry), dir); +return __fscrypt_prepare_link (d_inode (old_dentry), dir, dentry);return 0 ; }
include/linux/fscrypt_notsupp.h +2 -2
@@ -183,8 +183,8 @@ static inline int fscrypt_file_open(strureturn 0 ; } -static inline int __fscrypt_prepare_link (struct inode *inode, -struct inode *dir) +static inline int __fscrypt_prepare_link (struct inode *inode,struct inode *dir, +struct dentry *dentry) {return -EOPNOTSUPP ; }
include/linux/fscrypt_supp.h +2 -1
@@ -184,7 +184,8 @@ extern int fscrypt_zeroout_range(const s/* hooks.c */ extern int fscrypt_file_open (struct inode *inode,struct file *filp); -extern int __fscrypt_prepare_link (struct inode *inode,struct inode *dir); +extern int __fscrypt_prepare_link (struct inode *inode,struct inode *dir, +struct dentry *dentry); extern int __fscrypt_prepare_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir,
Greg Kroah-Hartman
maintainerPATCH 32/191 fs, fscrypt: clear DCACHE_ENCRYPTED_NAME when unaliasing directory
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit 0bf3d5c1604ecbbd4e49e9f5b3c79152b87adb0d upstream.
Make __d_move() clear DCACHE_ENCRYPTED_NAME on the source dentry. This is needed for when d_splice_alias() moves a directory's encrypted alias to its decrypted alias as a result of the encryption key being added.
Otherwise, the decrypted alias will incorrectly be invalidated on the next lookup, causing problems such as unmounting a mount the user just mount()ed there.
Note that we don't have to support arbitrary moves of this flag because fscrypt doesn't allow dentries with DCACHE_ENCRYPTED_NAME to be the source or target of a rename().
--- fs/dcache.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+)
fs/dcache.c +15 -0
@@ -2713,6 +2713,20 @@ static void copy_name(struct dentry *den }/* +* When d_splice_alias() moves a directory's encrypted alias to its decrypted +* alias as a result of the encryption key being added, DCACHE_ENCRYPTED_NAME +* must be cleared. Note that we don't have to support arbitrary moves of this +* flag because fscrypt doesn't allow encrypted aliases to be the source or +* target of a rename(). +*/ +static inline void fscrypt_handle_d_move (struct dentry *dentry) +{ +#if IS_ENABLED (CONFIG_FS_ENCRYPTION ) + dentry->d_flags &= ~DCACHE_ENCRYPTED_NAME ; +#endif +} + +/* * __d_move - move a dentry * @dentry: entry to move * @target: new dentry
@@ -2787,6 +2801,7 @@ static void __d_move(struct dentry *dentlist_move (&dentry->d_child, &dentry->d_parent->d_subdirs);__d_rehash (dentry);fsnotify_update_flags (dentry); +fscrypt_handle_d_move (dentry);write_seqcount_end (&target->d_seq);write_seqcount_end (&dentry->d_seq);
Greg Kroah-Hartman
maintainerPATCH 33/191 fscrypt: only set dentry_operations on ciphertext dentries
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit d456a33f041af4b54f3ce495a86d00c246165032 upstream.
Plaintext dentries are always valid, so only set fscrypt_d_ops on ciphertext dentries.
Besides marginally improved performance, this allows overlayfs to use an fscrypt-encrypted upperdir, provided that all the following are true:
(1) The fscrypt encryption key is placed in the keyring before mounting overlayfs, and remains while the overlayfs is mounted.
(2) The overlayfs workdir uses the same encryption policy.
(3) No dentries for the ciphertext names of subdirectories have been created in the upperdir or workdir yet. (Since otherwise d_splice_alias() will reuse the old dentry with ->d_op set.)
One potential use case is using an ephemeral encryption key to encrypt all files created or changed by a container, so that they can be securely erased ("crypto-shredded") after the container stops.
--- fs/crypto/hooks.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
fs/crypto/hooks.c +1 -2
@@ -115,9 +115,8 @@ int __fscrypt_prepare_lookup(struct inodspin_lock (&dentry->d_lock); dentry->d_flags |=DCACHE_ENCRYPTED_NAME ;spin_unlock (&dentry->d_lock); +d_set_d_op (dentry, &fscrypt_d_ops); } - -d_set_d_op (dentry, &fscrypt_d_ops);return 0 ; }EXPORT_SYMBOL_GPL (__fscrypt_prepare_lookup);
Greg Kroah-Hartman
maintainerPATCH 34/191 fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext
In reply to Greg Kroah-HartmanFrom: Eric Biggers <[email protected]>
commit b01531db6cec2aa330dbc91bfbfaaef4a0d387a4 upstream.
->lookup() in an encrypted directory begins as follows:
1. fscrypt_prepare_lookup():
a. Try to load the directory's encryption key.
b. If the key is unavailable, mark the dentry as a ciphertext name
via d_flags.
2. fscrypt_setup_filename():
a. Try to load the directory's encryption key.
b. If the key is available, encrypt the name (treated as a plaintext
name) to get the on-disk name. Otherwise decode the name (treated as a ciphertext name) to get the on-disk name.
But if the key is concurrently added, it may be found at (2a) but not at (1a). In this case, the dentry will be wrongly marked as a ciphertext name even though it was actually treated as plaintext.
This will cause the dentry to be wrongly invalidated on the next lookup, potentially causing problems. For example, if the racy ->lookup() was part of sys_mount(), then the new mount will be detached when anything tries to access it. This is despite the mountpoint having a plaintext path, which should remain valid now that the key was added.
Of course, this is only possible if there's a userspace race. Still, the additional kernel-side race is confusing and unexpected.
Close the kernel-side race by changing fscrypt_prepare_lookup() to also set the on-disk filename (step 2b), consistent with the d_flags update.
--- fs/crypto/fname.c | 1 fs/crypto/hooks.c | 11 +++-- fs/ext4/ext4.h | 62 ++++++++++++++++++++++++-------- fs/ext4/namei.c | 76 +++++++++++++++++++++++++++------------- fs/f2fs/namei.c | 17 +++++--- fs/ubifs/dir.c | 8 +--- include/linux/fscrypt.h | 22 +++++++---- include/linux/fscrypt_notsupp.h | 5 +- include/linux/fscrypt_supp.h | 3 + 9 files changed, 139 insertions(+), 66 deletions(-)
fs/crypto/fname.c +1 -0
@@ -354,6 +354,7 @@ int fscrypt_setup_filename(struct inode }if (!lookup)return -ENOKEY ; + fname->is_ciphertext_name = true;/* * We don't have the key and we are doing a lookup; decode the
fs/crypto/hooks.c +6 -5
@@ -104,20 +104,21 @@ int __fscrypt_prepare_rename(struct inod }EXPORT_SYMBOL_GPL (__fscrypt_prepare_rename); -int __fscrypt_prepare_lookup (struct inode *dir,struct dentry *dentry) +int __fscrypt_prepare_lookup (struct inode *dir,struct dentry *dentry, +struct fscrypt_name *fname) { -int err =fscrypt_get_encryption_info (dir); +int err =fscrypt_setup_filename (dir, &dentry->d_name,1 , fname); -if (err) +if (err && err != -ENOENT )return err; -if (!fscrypt_has_encryption_key (dir)) { +if (fname->is_ciphertext_name) {spin_lock (&dentry->d_lock); dentry->d_flags |=DCACHE_ENCRYPTED_NAME ;spin_unlock (&dentry->d_lock);d_set_d_op (dentry, &fscrypt_d_ops); } -return 0 ; +return err; }EXPORT_SYMBOL_GPL (__fscrypt_prepare_lookup);
fs/ext4/ext4.h +47 -15
89 lines not on this page · raw message
fs/ext4/namei.c +52 -24
138 lines not on this page · raw message
fs/f2fs/namei.c +10 -7
@@ -432,19 +432,23 @@ static struct dentry *f2fs_lookup(structnid_t ino =-1 ;int err =0 ;unsigned int root_ino =F2FS_ROOT_INO (F2FS_I_SB (dir)); +struct fscrypt_name fname;trace_f2fs_lookup_start (dir, dentry, flags); - err =fscrypt_prepare_lookup (dir, dentry, flags); -if (err) - goto out; -if (dentry->d_name.len >F2FS_NAME_LEN ) { err = -ENAMETOOLONG ; goto out; } - de =f2fs_find_entry (dir, &dentry->d_name, &page); + err =fscrypt_prepare_lookup (dir, dentry, &fname); +if (err == -ENOENT ) + goto out_splice; +if (err) + goto out; + de =__f2fs_find_entry (dir, &fname, &page); +fscrypt_free_filename (&fname); +if (!de) {if (IS_ERR (page)) { err =PTR_ERR (page);
@@ -484,8 +488,7 @@ static struct dentry *f2fs_lookup(struct } out_splice: new =d_splice_alias (inode, dentry); -if (IS_ERR (new)) - err =PTR_ERR (new); + err =PTR_ERR_OR_ZERO (new);trace_f2fs_lookup_end (dir, dentry, ino, err);return new; out_iput:
fs/ubifs/dir.c +3 -5
@@ -220,11 +220,9 @@ static struct dentry *ubifs_lookup(strucdbg_gen ("'%pd' in dir ino %lu" , dentry, dir->i_ino); - err =fscrypt_prepare_lookup (dir, dentry, flags); -if (err) -return ERR_PTR (err); - - err =fscrypt_setup_filename (dir, &dentry->d_name,1 , &nm); + err =fscrypt_prepare_lookup (dir, dentry, &nm); +if (err == -ENOENT ) +return d_splice_alias (NULL , dentry);if (err)return ERR_PTR (err);
include/linux/fscrypt.h +15 -7
@@ -32,6 +32,7 @@ struct fscrypt_name {u32 hash;u32 minor_hash;struct fscrypt_str crypto_buf; +bool is_ciphertext_name; };#define FSTR_INIT (n, l) { .name = n, .len = l }
@@ -138,25 +139,32 @@ static inline int fscrypt_prepare_rename* fscrypt_prepare_lookup - prepare to lookup a name in a possibly-encrypted directory * @dir: directory being searched * @dentry: filename being looked up -* @flags: lookup flags +* @fname: (output) the name to use to search the on-disk directory * -* Prepare for ->lookup() in a directory which may be encrypted. Lookups can be -* done with or without the directory's encryption key; without the key, +* Prepare for ->lookup() in a directory which may be encrypted by determining +* the name that will actually be used to search the directory on-disk. Lookups +* can be done with or without the directory's encryption key; without the key, * filenames are presented in encrypted form. Therefore, we'll try to set up * the directory's encryption key, but even without it the lookup can continue. * * This also installs a custom ->d_revalidate() method which will invalidate the * dentry if it was created without the key and the key is later added. * -* Return: 0 on success, -errno if a problem occurred while setting up the -* encryption key +* Return: 0 on success; -ENOENT if key is unavailable but the filename isn't a +* correctly formed encoded ciphertext name, so a negative dentry should be +* created; or another -errno code. */ static inline int fscrypt_prepare_lookup (struct inode *dir,struct dentry *dentry, -unsigned int flags) +struct fscrypt_name *fname) {if (IS_ENCRYPTED (dir)) -return __fscrypt_prepare_lookup (dir, dentry); +return __fscrypt_prepare_lookup (dir, dentry, fname); + +memset (fname,0 ,sizeof (*fname)); + fname->usr_fname = &dentry->d_name; + fname->disk_name.name = (unsigned char *)dentry->d_name.name; + fname->disk_name.len = dentry->d_name.len;return 0 ; }
include/linux/fscrypt_notsupp.h +3 -2
@@ -112,7 +112,7 @@ static inline int fscrypt_setup_filenameif (IS_ENCRYPTED (dir))return -EOPNOTSUPP ; -memset (fname,0 ,sizeof (struct fscrypt_name )); +memset (fname,0 ,sizeof (*fname)); fname->usr_fname = iname; fname->disk_name.name = (unsigned char *)iname->name; fname->disk_name.len = iname->len;
@@ -199,7 +199,8 @@ static inline int __fscrypt_prepare_rena }static inline int __fscrypt_prepare_lookup (struct inode *dir, -struct dentry *dentry) +struct dentry *dentry, +struct fscrypt_name *fname) {return -EOPNOTSUPP ; }
include/linux/fscrypt_supp.h +2 -1
@@ -191,7 +191,8 @@ extern int __fscrypt_prepare_rename(strustruct inode *new_dir, struct dentry *new_dentry, unsigned int flags); -extern int __fscrypt_prepare_lookup (struct inode *dir,struct dentry *dentry); +extern int __fscrypt_prepare_lookup (struct inode *dir,struct dentry *dentry, +struct fscrypt_name *fname);extern int __fscrypt_prepare_symlink (struct inode *dir,unsigned int len,unsigned int max_len,struct fscrypt_str *disk_link);
Eric Biggers
maintainerRe: [PATCH 4.19 034/191] fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext
Looks like something went wrong with formatting the email addresses that receive these emails. The Cc line has:
Cc: Greg Kroah-Hartman <[email protected]>, "[email protected], [email protected],
[email protected], [email protected], Theodore Tso" <[email protected]>, Eric
Biggers <[email protected]>, Theodore Ts'o <[email protected]>
The list addresses are part of display name of "[email protected]", so they apparently didn't receive this email.
- Eric
Greg Kroah-Hartman
maintainerRe: [PATCH 4.19 034/191] fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext
Quoted 9 lines · Eric Biggers
Looks like something went wrong with formatting the email addresses that receive these emails. The Cc line has:
Cc: Greg Kroah-Hartman <[email protected]>, "[email protected], [email protected],
[email protected], [email protected], Theodore Tso" <[email protected]>, Eric
Biggers <[email protected]>, Theodore Ts'o <[email protected]>The list addresses are part of display name of "[email protected]", so they apparently didn't receive this email.
Quilt's handling of the ' character in email cc: is horrible, and caused this mistake, sorry. One of these days I'll figure out what needs to be patched to fix that up...
thanks,
greg k-h
Greg Kroah-Hartman
maintainerPATCH 35/191 Revert "block: ratelimit handle_bad_sector() message"
In reply to Greg Kroah-HartmanFrom: Greg Kroah-Hartman <[email protected]>
This reverts commit f86b9bf6228bb334fe1addcd566a658ecbd08f7e which is commit f4ac712e4fe009635344b9af5d890fe25fcc8c0d upstream.
Jari Ruusu writes:
Above change "block: ratelimit handle_bad_sector() message" upstream commit f4ac712e4fe009635344b9af5d890fe25fcc8c0d in 4.19.154 kernel is not completely OK.
Removing casts from arguments 4 and 5 produces these compile warnings:
...
For 64 bit systems it is only compile time cosmetic warning. For 32 bit system + CONFIG_LBDAF=n it introduces bugs: output formats are "%llu" and passed parameters are 32 bits. That is not OK.
Upstream kernels have hardcoded 64 bit sector_t. In older stable trees sector_t can be either 64 or 32 bit. In other words, backport of above patch needs to keep those original casts.
And Tetsuo Handa writes:
Indeed, commit f4ac712e4fe00963 ("block: ratelimit handle_bad_sector() message") depends on commit 72deb455b5ec619f ("block: remove CONFIG_LBDAF") which was merged into 5.2 kernel.
So let's revert it.
--- block/blk-core.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
block/blk-core.c +5 -4
@@ -2127,10 +2127,11 @@ static void handle_bad_sector(struct bio {char b[BDEVNAME_SIZE ]; -pr_info_ratelimited ("attempt to access beyond end of device\n" -"%s: rw=%d, want=%llu, limit=%llu\n" , -bio_devname (bio, b), bio->bi_opf, -bio_end_sector (bio), maxsector); +printk (KERN_INFO "attempt to access beyond end of device\n" ); +printk (KERN_INFO "%s: rw=%d, want=%Lu, limit=%Lu\n" , +bio_devname (bio, b), bio->bi_opf, + (unsigned long long )bio_end_sector (bio), + (long long )maxsector); }#ifdef CONFIG_FAIL_MAKE_REQUEST
Greg Kroah-Hartman
maintainerPATCH 36/191 xen/events: dont use chip_data for legacy IRQs
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 0891fb39ba67bd7ae023ea0d367297ffff010781 upstream.
Since commit c330fb1ddc0a ("XEN uses irqdesc::irq_data_common::handler_data to store a per interrupt XEN data pointer which contains XEN specific information.") Xen is using the chip_data pointer for storing IRQ specific data. When
running as a HVM domain this can result in problems for legacy IRQs, as
those might use chip_data for their own purposes.
Use a local array for this purpose in case of legacy IRQs, avoiding the double use.
--- drivers/xen/events/events_base.c | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-)
drivers/xen/events/events_base.c +21 -8
@@ -90,6 +90,8 @@ static bool (*pirq_needs_eoi)(unsigned i/* Xen will never allocate port zero for any purpose. */ #define VALID_EVTCHN (chn) ((chn) != 0) +static struct irq_info *legacy_info_ptrs[NR_IRQS_LEGACY ]; +static struct irq_chip xen_dynamic_chip;static struct irq_chip xen_percpu_chip;static struct irq_chip xen_pirq_chip;
@@ -154,7 +156,18 @@ int get_evtchn_to_irq(unsigned evtchn)/* Get info for IRQ */ struct irq_info *info_for_irq (unsigned irq) { -return irq_get_chip_data (irq); +if (irq <nr_legacy_irqs ()) +return legacy_info_ptrs[irq]; +else +return irq_get_chip_data (irq); +} + +static void set_info_for_irq (unsigned int irq,struct irq_info *info) +{ +if (irq <nr_legacy_irqs ()) + legacy_info_ptrs[irq] = info; +else +irq_set_chip_data (irq, info); }/* Constructors for packed IRQ information. */
@@ -375,7 +388,7 @@ static void xen_irq_init(unsigned irq) info->type =IRQT_UNBOUND ; info->refcnt =-1 ; -irq_set_chip_data (irq, info); +set_info_for_irq (irq, info);list_add_tail (&info->list, &xen_irq_list_head); }
@@ -424,14 +437,14 @@ static int __must_check xen_allocate_irqstatic void xen_free_irq (unsigned irq) { -struct irq_info *info =irq_get_chip_data (irq); +struct irq_info *info =info_for_irq (irq);if (WARN_ON (!info))return ;list_del (&info->list); -irq_set_chip_data (irq,NULL ); +set_info_for_irq (irq,NULL );WARN_ON (info->refcnt >0 );
@@ -601,7 +614,7 @@ EXPORT_SYMBOL_GPL(xen_irq_from_gsi);static void __unbind_from_irq (unsigned int irq) {int evtchn =evtchn_from_irq (irq); -struct irq_info *info =irq_get_chip_data (irq); +struct irq_info *info =info_for_irq (irq);if (info->refcnt >0 ) { info->refcnt--;
@@ -1105,7 +1118,7 @@ int bind_ipi_to_irqhandler(enum ipi_vectvoid unbind_from_irqhandler (unsigned int irq,void *dev_id) { -struct irq_info *info =irq_get_chip_data (irq); +struct irq_info *info =info_for_irq (irq);if (WARN_ON (!info))return ;
@@ -1139,7 +1152,7 @@ int evtchn_make_refcounted(unsigned intif (irq ==-1 )return -ENOENT ; - info =irq_get_chip_data (irq); + info =info_for_irq (irq);if (!info)return -ENOENT ;
@@ -1167,7 +1180,7 @@ int evtchn_get(unsigned int evtchn)if (irq ==-1 ) goto done; - info =irq_get_chip_data (irq); + info =info_for_irq (irq);if (!info) goto done;
Greg Kroah-Hartman
maintainerPATCH 37/191 xen/events: avoid removing an event channel while handling it
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 073d0552ead5bfc7a3a9c01de590e924f11b5dd2 upstream.
Today it can happen that an event channel is being removed from the system while the event handling loop is active. This can lead to a race resulting in crashes or WARN() splats when trying to access the irq_info structure related to the event channel.
Fix this problem by using a rwlock taken as reader in the event handling loop and as writer when deallocating the irq_info structure.
As the observed problem was a NULL dereference in evtchn_from_irq() make this function more robust against races by testing the irq_info pointer to be not NULL before dereferencing it.
And finally make all accesses to evtchn_to_irq[row][col] atomic ones in order to avoid seeing partial updates of an array element in irq handling. Note that irq handling can be entered only for event channels which have been valid before, so any not populated row isn't a problem in this regard, as rows are only ever added and never removed.
This is XSA-331.
--- drivers/xen/events/events_base.c | 40 ++++++++++++++++++++++++++++++++++----- 1 file changed, 35 insertions(+), 5 deletions(-)
drivers/xen/events/events_base.c +35 -5
@@ -32,6 +32,7 @@#include <linux/slab.h> #include <linux/irqnr.h> #include <linux/pci.h> +#include <linux/spinlock.h> #ifdef CONFIG_X86 #include <asm/desc.h>
@@ -69,6 +70,23 @@ const struct evtchn_ops *evtchn_ops;*/ static DEFINE_MUTEX (irq_mapping_update_lock); +/* +* Lock protecting event handling loop against removing event channels. +* Adding of event channels is no issue as the associated IRQ becomes active +* only after everything is setup (before request_[threaded_]irq() the handler +* can't be entered for an event, as the event channel will be unmasked only +* then). +*/ +static DEFINE_RWLOCK (evtchn_rwlock); + +/* +* Lock hierarchy: +* +* irq_mapping_update_lock +* evtchn_rwlock +* IRQ-desc lock +*/ +static LIST_HEAD (xen_irq_list_head);/* IRQ <-> VIRQ mapping. */
@@ -103,7 +121,7 @@ static void clear_evtchn_to_irq_row(unsiunsigned col;for (col =0 ; col <EVTCHN_PER_ROW ; col++) - evtchn_to_irq[row][col] =-1 ; +WRITE_ONCE (evtchn_to_irq[row][col],-1 ); }static void clear_evtchn_to_irq_all (void )
@@ -140,7 +158,7 @@ static int set_evtchn_to_irq(unsigned evclear_evtchn_to_irq_row (row); } - evtchn_to_irq[row][col] = irq; +WRITE_ONCE (evtchn_to_irq[row][col], irq);return 0 ; }
@@ -150,7 +168,7 @@ int get_evtchn_to_irq(unsigned evtchn)return -1 ;if (evtchn_to_irq[EVTCHN_ROW (evtchn)] ==NULL )return -1 ; -return evtchn_to_irq[EVTCHN_ROW (evtchn)][EVTCHN_COL (evtchn)]; +return READ_ONCE (evtchn_to_irq[EVTCHN_ROW (evtchn)][EVTCHN_COL (evtchn)]); }/* Get info for IRQ */
@@ -259,10 +277,14 @@ static void xen_irq_info_cleanup(struct*/ unsigned int evtchn_from_irq (unsigned irq) { -if (unlikely (WARN (irq >= nr_irqs,"Invalid irq %d!\n" , irq))) +const struct irq_info *info =NULL ; + +if (likely (irq < nr_irqs)) + info =info_for_irq (irq); +if (!info)return 0 ; -return info_for_irq (irq)->evtchn; +return info->evtchn; }unsigned irq_from_evtchn (unsigned int evtchn)
@@ -438,16 +460,21 @@ static int __must_check xen_allocate_irqstatic void xen_free_irq (unsigned irq) {struct irq_info *info =info_for_irq (irq); +unsigned long flags;if (WARN_ON (!info))return ; +write_lock_irqsave (&evtchn_rwlock, flags); +list_del (&info->list);set_info_for_irq (irq,NULL );WARN_ON (info->refcnt >0 ); +write_unlock_irqrestore (&evtchn_rwlock, flags); +kfree (info);/* Legacy IRQ descriptors are managed by the arch. */
@@ -1233,6 +1260,8 @@ static void __xen_evtchn_do_upcall(void)int cpu =get_cpu ();unsigned count; +read_lock (&evtchn_rwlock); + do { vcpu_info->evtchn_upcall_pending = 0;
@@ -1248,6 +1277,7 @@ static void __xen_evtchn_do_upcall(void) }while (count !=1 || vcpu_info->evtchn_upcall_pending); out: +read_unlock (&evtchn_rwlock);put_cpu (); }
Greg Kroah-Hartman
maintainerPATCH 38/191 xen/events: add a proper barrier to 2-level uevent unmasking
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 4d3fe31bd993ef504350989786858aefdb877daa upstream.
A follow-up patch will require certain write to happen before an event channel is unmasked.
While the memory barrier is not strictly necessary for all the callers, the main one will need it. In order to avoid an extra memory barrier when using fifo event channels, mandate evtchn_unmask() to provide write ordering.
The 2-level event handling unmask operation is missing an appropriate barrier, so add it. Fifo event channels are fine in this regard due to using sync_cmpxchg().
This is part of XSA-332.
--- drivers/xen/events/events_2l.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/xen/events/events_2l.c +2 -0
@@ -91,6 +91,8 @@ static void evtchn_2l_unmask(unsigned poBUG_ON (!irqs_disabled ()); +smp_wmb ();/* All writes before unmask must be visible. */ +if (unlikely ((cpu !=cpu_from_evtchn (port)))) do_hypercall =1 ;else {
Greg Kroah-Hartman
maintainerPATCH 39/191 xen/events: fix race in evtchn_fifo_unmask()
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit f01337197419b7e8a492e83089552b77d3b5fb90 upstream.
Unmasking a fifo event channel can result in unmasking it twice, once directly in the kernel and once via a hypercall in case the event was pending.
Fix that by doing the local unmask only if the event is not pending.
This is part of XSA-332.
--- drivers/xen/events/events_fifo.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/xen/events/events_fifo.c +9 -4
@@ -227,19 +227,25 @@ static bool evtchn_fifo_is_masked(unsignreturn sync_test_bit (EVTCHN_FIFO_BIT (MASKED , word),BM (word)); }/* -* Clear MASKED, spinning if BUSY is set. +* Clear MASKED if not PENDING, spinning if BUSY is set. +* Return true if mask was cleared. */ -static void clear_masked (volatile event_word_t *word) +static bool clear_masked_cond (volatile event_word_t *word) {event_word_t new, old, w; w = *word;do { +if (w & (1 <<EVTCHN_FIFO_PENDING )) +return false; + old = w & ~(1 <<EVTCHN_FIFO_BUSY ); new = old & ~(1 <<EVTCHN_FIFO_MASKED ); w =sync_cmpxchg (word, old, new); }while (w != old); + +return true; }static void evtchn_fifo_unmask (unsigned port)
@@ -248,8 +254,7 @@ static void evtchn_fifo_unmask(unsignedBUG_ON (!irqs_disabled ()); -clear_masked (word); -if (evtchn_fifo_is_pending (port)) { +if (!clear_masked_cond (word)) {struct evtchn_unmask unmask = { .port = port }; (void )HYPERVISOR_event_channel_op (EVTCHNOP_unmask, &unmask); }
Greg Kroah-Hartman
maintainerPATCH 40/191 xen/events: add a new "late EOI" evtchn framework
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 54c9de89895e0a36047fcc4ae754ea5b8655fb9d upstream.
In order to avoid tight event channel related IRQ loops add a new framework of "late EOI" handling: the IRQ the event channel is bound to will be masked until the event has been handled and the related driver is capable to handle another event. The driver is responsible for unmasking the event channel via the new function xen_irq_lateeoi().
This is similar to binding an event channel to a threaded IRQ, but without having to structure the driver accordingly.
In order to support a future special handling in case a rogue guest is sending lots of unsolicited events, add a flag to xen_irq_lateeoi() which can be set by the caller to indicate the event was a spurious one.
This is part of XSA-332.
--- drivers/xen/events/events_base.c | 151 ++++++++++++++++++++++++++++++++++----- include/xen/events.h | 29 ++++++- 2 files changed, 159 insertions(+), 21 deletions(-)
drivers/xen/events/events_base.c +134 -17
@@ -111,6 +111,7 @@ static bool (*pirq_needs_eoi)(unsigned istatic struct irq_info *legacy_info_ptrs[NR_IRQS_LEGACY ];static struct irq_chip xen_dynamic_chip; +static struct irq_chip xen_lateeoi_chip;static struct irq_chip xen_percpu_chip;static struct irq_chip xen_pirq_chip;static void enable_dynirq (struct irq_data *data);
@@ -395,6 +396,33 @@ void notify_remote_via_irq(int irq) }EXPORT_SYMBOL_GPL (notify_remote_via_irq); +static void xen_irq_lateeoi_locked (struct irq_info *info) +{ +evtchn_port_t evtchn; + + evtchn = info->evtchn; +if (!VALID_EVTCHN (evtchn)) +return ; + +unmask_evtchn (evtchn); +} + +void xen_irq_lateeoi (unsigned int irq,unsigned int eoi_flags) +{ +struct irq_info *info; +unsigned long flags; + +read_lock_irqsave (&evtchn_rwlock, flags); + + info =info_for_irq (irq); + +if (info) +xen_irq_lateeoi_locked (info); + +read_unlock_irqrestore (&evtchn_rwlock, flags); +} +EXPORT_SYMBOL_GPL (xen_irq_lateeoi); +static void xen_irq_init (unsigned irq) {struct irq_info *info;
@@ -866,7 +894,7 @@ int xen_pirq_from_irq(unsigned irq) }EXPORT_SYMBOL_GPL (xen_pirq_from_irq); -int bind_evtchn_to_irq (unsigned int evtchn) +static int bind_evtchn_to_irq_chip (evtchn_port_t evtchn,struct irq_chip *chip) {int irq;int ret;
@@ -883,7 +911,7 @@ int bind_evtchn_to_irq(unsigned int evtcif (irq <0 ) goto out; -irq_set_chip_and_handler_name (irq, &xen_dynamic_chip, +irq_set_chip_and_handler_name (irq, chip, handle_edge_irq,"event" ); ret =xen_irq_info_evtchn_setup (irq, evtchn);
@@ -904,8 +932,19 @@ out:return irq; } + +int bind_evtchn_to_irq (evtchn_port_t evtchn) +{ +return bind_evtchn_to_irq_chip (evtchn, &xen_dynamic_chip); +}EXPORT_SYMBOL_GPL (bind_evtchn_to_irq); +int bind_evtchn_to_irq_lateeoi (evtchn_port_t evtchn) +{ +return bind_evtchn_to_irq_chip (evtchn, &xen_lateeoi_chip); +} +EXPORT_SYMBOL_GPL (bind_evtchn_to_irq_lateeoi); +static int bind_ipi_to_irq (unsigned int ipi,unsigned int cpu) {struct evtchn_bind_ipi bind_ipi;
@@ -947,8 +986,9 @@ static int bind_ipi_to_irq(unsigned intreturn irq; } -int bind_interdomain_evtchn_to_irq (unsigned int remote_domain, -unsigned int remote_port) +static int bind_interdomain_evtchn_to_irq_chip (unsigned int remote_domain, +evtchn_port_t remote_port, +struct irq_chip *chip) {struct evtchn_bind_interdomain bind_interdomain;int err;
@@ -959,10 +999,26 @@ int bind_interdomain_evtchn_to_irq(unsig err =HYPERVISOR_event_channel_op (EVTCHNOP_bind_interdomain, &bind_interdomain); -return err ? :bind_evtchn_to_irq (bind_interdomain.local_port); +return err ? :bind_evtchn_to_irq_chip (bind_interdomain.local_port, + chip); +} + +int bind_interdomain_evtchn_to_irq (unsigned int remote_domain, +evtchn_port_t remote_port) +{ +return bind_interdomain_evtchn_to_irq_chip (remote_domain, remote_port, + &xen_dynamic_chip); }EXPORT_SYMBOL_GPL (bind_interdomain_evtchn_to_irq); +int bind_interdomain_evtchn_to_irq_lateeoi (unsigned int remote_domain, +evtchn_port_t remote_port) +{ +return bind_interdomain_evtchn_to_irq_chip (remote_domain, remote_port, + &xen_lateeoi_chip); +} +EXPORT_SYMBOL_GPL (bind_interdomain_evtchn_to_irq_lateeoi); +static int find_virq (unsigned int virq,unsigned int cpu) {struct evtchn_status status;
@@ -1058,14 +1114,15 @@ static void unbind_from_irq(unsigned intmutex_unlock (&irq_mapping_update_lock); } -int bind_evtchn_to_irqhandler (unsigned int evtchn, -irq_handler_t handler, -unsigned long irqflags, -const char *devname,void *dev_id) +static int bind_evtchn_to_irqhandler_chip (evtchn_port_t evtchn, +irq_handler_t handler, +unsigned long irqflags, +const char *devname,void *dev_id, +struct irq_chip *chip) {int irq, retval; - irq =bind_evtchn_to_irq (evtchn); + irq =bind_evtchn_to_irq_chip (evtchn, chip);if (irq <0 )return irq; retval =request_irq (irq, handler, irqflags, devname, dev_id);
@@ -1076,18 +1133,38 @@ int bind_evtchn_to_irqhandler(unsigned ireturn irq; } + +int bind_evtchn_to_irqhandler (evtchn_port_t evtchn, +irq_handler_t handler, +unsigned long irqflags, +const char *devname,void *dev_id) +{ +return bind_evtchn_to_irqhandler_chip (evtchn, handler, irqflags, + devname, dev_id, + &xen_dynamic_chip); +}EXPORT_SYMBOL_GPL (bind_evtchn_to_irqhandler); -int bind_interdomain_evtchn_to_irqhandler (unsigned int remote_domain, -unsigned int remote_port, -irq_handler_t handler, -unsigned long irqflags, -const char *devname, -void *dev_id) +int bind_evtchn_to_irqhandler_lateeoi (evtchn_port_t evtchn, +irq_handler_t handler, +unsigned long irqflags, +const char *devname,void *dev_id) +{ +return bind_evtchn_to_irqhandler_chip (evtchn, handler, irqflags, + devname, dev_id, + &xen_lateeoi_chip); +} +EXPORT_SYMBOL_GPL (bind_evtchn_to_irqhandler_lateeoi); + +static int bind_interdomain_evtchn_to_irqhandler_chip ( +unsigned int remote_domain,evtchn_port_t remote_port, +irq_handler_t handler,unsigned long irqflags, +const char *devname,void *dev_id,struct irq_chip *chip) {int irq, retval; - irq =bind_interdomain_evtchn_to_irq (remote_domain, remote_port); + irq =bind_interdomain_evtchn_to_irq_chip (remote_domain, remote_port, + chip);if (irq <0 )return irq;
@@ -1099,8 +1176,33 @@ int bind_interdomain_evtchn_to_irqhandlereturn irq; } + +int bind_interdomain_evtchn_to_irqhandler (unsigned int remote_domain, +evtchn_port_t remote_port, +irq_handler_t handler, +unsigned long irqflags, +const char *devname, +void *dev_id) +{ +return bind_interdomain_evtchn_to_irqhandler_chip (remote_domain, + remote_port, handler, irqflags, devname, + dev_id, &xen_dynamic_chip); +}EXPORT_SYMBOL_GPL (bind_interdomain_evtchn_to_irqhandler); +int bind_interdomain_evtchn_to_irqhandler_lateeoi (unsigned int remote_domain, +evtchn_port_t remote_port, +irq_handler_t handler, +unsigned long irqflags, +const char *devname, +void *dev_id) +{ +return bind_interdomain_evtchn_to_irqhandler_chip (remote_domain, + remote_port, handler, irqflags, devname, + dev_id, &xen_lateeoi_chip); +} +EXPORT_SYMBOL_GPL (bind_interdomain_evtchn_to_irqhandler_lateeoi); +int bind_virq_to_irqhandler (unsigned int virq,unsigned int cpu,irq_handler_t handler,unsigned long irqflags,const char *devname,void *dev_id)
@@ -1641,6 +1743,21 @@ static struct irq_chip xen_dynamic_chip .irq_mask_ack = mask_ack_dynirq, .irq_set_affinity = set_affinity_irq, + .irq_retrigger = retrigger_dynirq, +}; + +static struct irq_chip xen_lateeoi_chip__read_mostly = { +/* The chip name needs to contain "xen-dyn" for irqbalance to work. */ + .name ="xen-dyn-lateeoi" , + + .irq_disable = disable_dynirq, + .irq_mask = disable_dynirq, + .irq_unmask = enable_dynirq, + + .irq_ack = mask_ack_dynirq, + .irq_mask_ack = mask_ack_dynirq, + + .irq_set_affinity = set_affinity_irq, .irq_retrigger = retrigger_dynirq, };
include/xen/events.h +25 -4
@@ -14,11 +14,16 @@unsigned xen_evtchn_nr_channels (void ); -int bind_evtchn_to_irq (unsigned int evtchn); -int bind_evtchn_to_irqhandler (unsigned int evtchn, +int bind_evtchn_to_irq (evtchn_port_t evtchn); +int bind_evtchn_to_irq_lateeoi (evtchn_port_t evtchn); +int bind_evtchn_to_irqhandler (evtchn_port_t evtchn,irq_handler_t handler,unsigned long irqflags,const char *devname,void *dev_id); +int bind_evtchn_to_irqhandler_lateeoi (evtchn_port_t evtchn, +irq_handler_t handler, +unsigned long irqflags,const char *devname, +void *dev_id);int bind_virq_to_irq (unsigned int virq,unsigned int cpu,bool percpu); int bind_virq_to_irqhandler(unsigned int virq, unsigned int cpu, irq_handler_t handler,
@@ -31,13 +36,21 @@ int bind_ipi_to_irqhandler(enum ipi_vectconst char *devname, void *dev_id);int bind_interdomain_evtchn_to_irq (unsigned int remote_domain, -unsigned int remote_port); +evtchn_port_t remote_port); +int bind_interdomain_evtchn_to_irq_lateeoi (unsigned int remote_domain, +evtchn_port_t remote_port);int bind_interdomain_evtchn_to_irqhandler (unsigned int remote_domain, -unsigned int remote_port, +evtchn_port_t remote_port,irq_handler_t handler,unsigned long irqflags,const char *devname,void *dev_id); +int bind_interdomain_evtchn_to_irqhandler_lateeoi (unsigned int remote_domain, +evtchn_port_t remote_port, +irq_handler_t handler, +unsigned long irqflags, +const char *devname, +void *dev_id);/* * Common unbind function for all event sources. Takes IRQ to unbind from.
@@ -46,6 +59,14 @@ int bind_interdomain_evtchn_to_irqhandle*/ void unbind_from_irqhandler (unsigned int irq,void *dev_id); +/* +* Send late EOI for an IRQ bound to an event channel via one of the *_lateeoi +* functions above. +*/ +void xen_irq_lateeoi (unsigned int irq,unsigned int eoi_flags); +/* Signal an event was spurious, i.e. there was no action resulting from it. */ +#define XEN_EOI_FLAG_SPURIOUS 0x00000001 +#define XEN_IRQ_PRIORITY_MAX EVTCHN_FIFO_PRIORITY_MAX#define XEN_IRQ_PRIORITY_DEFAULT EVTCHN_FIFO_PRIORITY_DEFAULT#define XEN_IRQ_PRIORITY_MIN EVTCHN_FIFO_PRIORITY_MIN
Greg Kroah-Hartman
maintainerPATCH 41/191 xen/blkback: use lateeoi irq binding
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 01263a1fabe30b4d542f34c7e2364a22587ddaf2 upstream.
In order to reduce the chance for the system becoming unresponsive due to event storms triggered by a misbehaving blkfront use the lateeoi irq binding for blkback and unmask the event channel only after processing all pending requests.
As the thread processing requests is used to do purging work in regular intervals an EOI may be sent only after having received an event. If there was no pending I/O request flag the EOI as spurious.
This is part of XSA-332.
--- drivers/block/xen-blkback/blkback.c | 22 +++++++++++++++++----- drivers/block/xen-blkback/xenbus.c | 5 ++--- 2 files changed, 19 insertions(+), 8 deletions(-)
drivers/block/xen-blkback/blkback.c +17 -5
@@ -202,7 +202,7 @@ static inline void shrink_free_pagepool(#define vaddr (page) ((unsigned long)pfn_to_kaddr(page_to_pfn(page))) -static int do_block_io_op (struct xen_blkif_ring *ring); +static int do_block_io_op (struct xen_blkif_ring *ring,unsigned int *eoi_flags);static int dispatch_rw_block_io (struct xen_blkif_ring *ring,struct blkif_request *req,struct pending_req *pending_req);
@@ -615,6 +615,8 @@ int xen_blkif_schedule(void *arg)struct xen_vbd *vbd = &blkif->vbd;unsigned long timeout;int ret; +bool do_eoi; +unsigned int eoi_flags =XEN_EOI_FLAG_SPURIOUS ;set_freezable ();while (!kthread_should_stop ()) {
@@ -639,16 +641,23 @@ int xen_blkif_schedule(void *arg)if (timeout ==0 ) goto purge_gnt_list; + do_eoi = ring->waiting_reqs; + ring->waiting_reqs =0 ;smp_mb ();/* clear flag *before* checking for work */ - ret =do_block_io_op (ring); + ret =do_block_io_op (ring, &eoi_flags);if (ret >0 ) ring->waiting_reqs =1 ;if (ret == -EACCES )wait_event_interruptible (ring->shutdown_wq,kthread_should_stop ()); +if (do_eoi && !ring->waiting_reqs) { +xen_irq_lateeoi (ring->irq, eoi_flags); + eoi_flags |=XEN_EOI_FLAG_SPURIOUS ; + } + purge_gnt_list:if (blkif->vbd.feature_gnt_persistent &&time_after (jiffies, ring->next_lru)) {
@@ -1121,7 +1130,7 @@ static void end_block_io_op(struct bio ** and transmute it to the block API to hand it over to the proper block disk. */ static int -__do_block_io_op (struct xen_blkif_ring *ring) +__do_block_io_op (struct xen_blkif_ring *ring,unsigned int *eoi_flags) {union blkif_back_rings *blk_rings = &ring->blk_rings;struct blkif_request req;
@@ -1144,6 +1153,9 @@ __do_block_io_op(struct xen_blkif_ring *if (RING_REQUEST_CONS_OVERFLOW (&blk_rings->common, rc))break ; +/* We've seen a request, so clear spurious eoi flag. */ + *eoi_flags &= ~XEN_EOI_FLAG_SPURIOUS ; +if (kthread_should_stop ()) { more_to_do =1 ;break ;
@@ -1202,13 +1214,13 @@ done: }static int -do_block_io_op (struct xen_blkif_ring *ring) +do_block_io_op (struct xen_blkif_ring *ring,unsigned int *eoi_flags) {union blkif_back_rings *blk_rings = &ring->blk_rings;int more_to_do; do { - more_to_do = __do_block_io_op(ring); + more_to_do = __do_block_io_op(ring, eoi_flags); if (more_to_do) break;
drivers/block/xen-blkback/xenbus.c +2 -3
@@ -237,9 +237,8 @@ static int xen_blkif_map(struct xen_blkiBUG (); } - err =bind_interdomain_evtchn_to_irqhandler (blkif->domid, evtchn, - xen_blkif_be_int,0 , -"blkif-backend" , ring); + err =bind_interdomain_evtchn_to_irqhandler_lateeoi (blkif->domid, + evtchn, xen_blkif_be_int,0 ,"blkif-backend" , ring);if (err <0 ) {xenbus_unmap_ring_vfree (blkif->be->dev, ring->blk_ring); ring->blk_rings.common.sring =NULL ;
Greg Kroah-Hartman
maintainerPATCH 42/191 xen/netback: use lateeoi irq binding
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 23025393dbeb3b8b3b60ebfa724cdae384992e27 upstream.
In order to reduce the chance for the system becoming unresponsive due to event storms triggered by a misbehaving netfront use the lateeoi irq binding for netback and unmask the event channel only just before going to sleep waiting for new events.
Make sure not to issue an EOI when none is pending by introducing an eoi_pending element to struct xenvif_queue.
When no request has been consumed set the spurious flag when sending the EOI for an interrupt.
This is part of XSA-332.
--- drivers/net/xen-netback/common.h | 15 ++++++++ drivers/net/xen-netback/interface.c | 61 ++++++++++++++++++++++++++++++------ drivers/net/xen-netback/netback.c | 11 +++++- drivers/net/xen-netback/rx.c | 13 +++++-- 4 files changed, 86 insertions(+), 14 deletions(-)
drivers/net/xen-netback/common.h +15 -0
@@ -140,6 +140,20 @@ struct xenvif_queue { /* Per-queue datachar name[QUEUE_NAME_SIZE ];/* DEVNAME-qN */ struct xenvif *vif;/* Parent VIF */ +/* +* TX/RX common EOI handling. +* When feature-split-event-channels = 0, interrupt handler sets +* NETBK_COMMON_EOI, otherwise NETBK_RX_EOI and NETBK_TX_EOI are set +* by the RX and TX interrupt handlers. +* RX and TX handler threads will issue an EOI when either +* NETBK_COMMON_EOI or their specific bits (NETBK_RX_EOI or +* NETBK_TX_EOI) are set and they will reset those bits. +*/ +atomic_t eoi_pending; +#define NETBK_RX_EOI 0x01 +#define NETBK_TX_EOI 0x02 +#define NETBK_COMMON_EOI 0x04 +/* Use NAPI for guest TX */ struct napi_struct napi;/* When feature-split-event-channels = 0, tx_irq = rx_irq. */
@@ -357,6 +371,7 @@ int xenvif_dealloc_kthread(void *data);irqreturn_t xenvif_ctrl_irq_fn (int irq,void *data); +bool xenvif_have_rx_work (struct xenvif_queue *queue,bool test_kthread);void xenvif_rx_action (struct xenvif_queue *queue);void xenvif_rx_queue_tail (struct xenvif_queue *queue,struct sk_buff *skb);
drivers/net/xen-netback/interface.c +52 -9
111 lines not on this page · raw message
drivers/net/xen-netback/netback.c +10 -1
@@ -162,6 +162,10 @@ void xenvif_napi_schedule_or_enable_evenif (more_to_do)napi_schedule (&queue->napi); +else if (atomic_fetch_andnot (NETBK_TX_EOI |NETBK_COMMON_EOI , + &queue->eoi_pending) & + (NETBK_TX_EOI |NETBK_COMMON_EOI )) +xen_irq_lateeoi (queue->tx_irq,0 ); }static void tx_add_credit (struct xenvif_queue *queue)
@@ -1613,9 +1617,14 @@ static bool xenvif_ctrl_work_todo(structirqreturn_t xenvif_ctrl_irq_fn (int irq,void *data) {struct xenvif *vif = data; +unsigned int eoi_flag =XEN_EOI_FLAG_SPURIOUS ; -while (xenvif_ctrl_work_todo (vif)) +while (xenvif_ctrl_work_todo (vif)) {xenvif_ctrl_action (vif); + eoi_flag =0 ; + } + +xen_irq_lateeoi (irq, eoi_flag);return IRQ_HANDLED ; }
drivers/net/xen-netback/rx.c +9 -4
@@ -490,13 +490,13 @@ static bool xenvif_rx_queue_ready(structreturn queue->stalled && prod - cons >=1 ; } -static bool xenvif_have_rx_work (struct xenvif_queue *queue) +bool xenvif_have_rx_work (struct xenvif_queue *queue,bool test_kthread) {return xenvif_rx_ring_slots_available (queue) || (queue->vif->stall_timeout && (xenvif_rx_queue_stalled (queue) ||xenvif_rx_queue_ready (queue))) || -kthread_should_stop () || + (test_kthread &&kthread_should_stop ()) || queue->vif->disabled; }
@@ -527,15 +527,20 @@ static void xenvif_wait_for_rx_work(stru {DEFINE_WAIT (wait); -if (xenvif_have_rx_work (queue)) +if (xenvif_have_rx_work (queue, true))return ;for (;;) {long ret;prepare_to_wait (&queue->wq, &wait,TASK_INTERRUPTIBLE ); -if (xenvif_have_rx_work (queue)) +if (xenvif_have_rx_work (queue, true))break ; +if (atomic_fetch_andnot (NETBK_RX_EOI |NETBK_COMMON_EOI , + &queue->eoi_pending) & + (NETBK_RX_EOI |NETBK_COMMON_EOI )) +xen_irq_lateeoi (queue->rx_irq,0 ); + ret =schedule_timeout (xenvif_rx_queue_timeout (queue));if (!ret)break ;
Greg Kroah-Hartman
maintainerPATCH 43/191 xen/scsiback: use lateeoi irq binding
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 86991b6e7ea6c613b7692f65106076943449b6b7 upstream.
In order to reduce the chance for the system becoming unresponsive due to event storms triggered by a misbehaving scsifront use the lateeoi irq binding for scsiback and unmask the event channel only just before leaving the event handling function.
In case of a ring protocol error don't issue an EOI in order to avoid the possibility to use that for producing an event storm. This at once will result in no further call of scsiback_irq_fn(), so the ring_error struct member can be dropped and scsiback_do_cmd_fn() can signal the protocol error via a negative return value.
This is part of XSA-332.
--- drivers/xen/xen-scsiback.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-)
drivers/xen/xen-scsiback.c +13 -10
@@ -91,7 +91,6 @@ struct vscsibk_info {unsigned int irq;struct vscsiif_back_ring ring; -int ring_error;spinlock_t ring_lock;atomic_t nr_unreplied_reqs;
@@ -722,7 +721,8 @@ static struct vscsibk_pend *prepare_pendreturn pending_req; } -static int scsiback_do_cmd_fn (struct vscsibk_info *info) +static int scsiback_do_cmd_fn (struct vscsibk_info *info, +unsigned int *eoi_flags) {struct vscsiif_back_ring *ring = &info->ring;struct vscsiif_request ring_req;
@@ -739,11 +739,12 @@ static int scsiback_do_cmd_fn(struct vsc rc = ring->rsp_prod_pvt;pr_warn ("Dom%d provided bogus ring requests (%#x - %#x = %u). Halting ring processing\n" , info->domid, rp, rc, rp - rc); - info->ring_error =1 ; -return 0 ; +return -EINVAL ; }while ((rc != rp)) { + *eoi_flags &= ~XEN_EOI_FLAG_SPURIOUS ; +if (RING_REQUEST_CONS_OVERFLOW (ring, rc))break ;
@@ -802,13 +803,16 @@ static int scsiback_do_cmd_fn(struct vscstatic irqreturn_t scsiback_irq_fn (int irq,void *dev_id) {struct vscsibk_info *info = dev_id; +int rc; +unsigned int eoi_flags =XEN_EOI_FLAG_SPURIOUS ; -if (info->ring_error) -return IRQ_HANDLED ; - -while (scsiback_do_cmd_fn (info)) +while ((rc =scsiback_do_cmd_fn (info, &eoi_flags)) >0 )cond_resched (); +/* In case of a ring error we keep the event channel masked. */ +if (!rc) +xen_irq_lateeoi (irq, eoi_flags); +return IRQ_HANDLED ; }
@@ -829,7 +833,7 @@ static int scsiback_init_sring(struct vs sring = (struct vscsiif_sring *)area;BACK_RING_INIT (&info->ring, sring,PAGE_SIZE ); - err =bind_interdomain_evtchn_to_irq (info->domid, evtchn); + err =bind_interdomain_evtchn_to_irq_lateeoi (info->domid, evtchn);if (err <0 ) goto unmap_page;
@@ -1252,7 +1256,6 @@ static int scsiback_probe(struct xenbus_ info->domid = dev->otherend_id;spin_lock_init (&info->ring_lock); - info->ring_error =0 ;atomic_set (&info->nr_unreplied_reqs,0 );init_waitqueue_head (&info->waiting_to_free); info->dev = dev;
Greg Kroah-Hartman
maintainerPATCH 44/191 xen/pvcallsback: use lateeoi irq binding
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit c8d647a326f06a39a8e5f0f1af946eacfa1835f8 upstream.
In order to reduce the chance for the system becoming unresponsive due to event storms triggered by a misbehaving pvcallsfront use the lateeoi irq binding for pvcallsback and unmask the event channel only after handling all write requests, which are the ones coming in via an irq.
This requires modifying the logic a little bit to not require an event for each write request, but to keep the ioworker running until no further data is found on the ring page to be processed.
This is part of XSA-332.
--- drivers/xen/pvcalls-back.c | 76 +++++++++++++++++++++++++++------------------ 1 file changed, 46 insertions(+), 30 deletions(-)
drivers/xen/pvcalls-back.c +46 -30
@@ -75,6 +75,7 @@ struct sock_mapping {atomic_t write;atomic_t io;atomic_t release; +atomic_t eoi;void (*saved_data_ready)(struct sock *sk);struct pvcalls_ioworker ioworker; };
@@ -96,7 +97,7 @@ static int pvcalls_back_release_active(sstruct pvcalls_fedata *fedata, struct sock_mapping *map); -static void pvcalls_conn_back_read (void *opaque) +static bool pvcalls_conn_back_read (void *opaque) {struct sock_mapping *map = (struct sock_mapping *)opaque;struct msghdr msg;
@@ -116,17 +117,17 @@ static void pvcalls_conn_back_read(voidvirt_mb ();if (error) -return ; +return false; size =pvcalls_queued (prod, cons, array_size);if (size >= array_size) -return ; +return false;spin_lock_irqsave (&map->sock->sk->sk_receive_queue.lock, flags);if (skb_queue_empty (&map->sock->sk->sk_receive_queue)) {atomic_set (&map->read,0 );spin_unlock_irqrestore (&map->sock->sk->sk_receive_queue.lock, flags); -return ; +return true; }spin_unlock_irqrestore (&map->sock->sk->sk_receive_queue.lock, flags); wanted = array_size - size;
@@ -150,7 +151,7 @@ static void pvcalls_conn_back_read(void ret =inet_recvmsg (map->sock, &msg, wanted,MSG_DONTWAIT );WARN_ON (ret > wanted);if (ret == -EAGAIN )/* shouldn't happen */ -return ; +return true;if (!ret) ret = -ENOTCONN ;spin_lock_irqsave (&map->sock->sk->sk_receive_queue.lock, flags);
@@ -169,10 +170,10 @@ static void pvcalls_conn_back_read(voidvirt_wmb ();notify_remote_via_irq (map->irq); -return ; +return true; } -static void pvcalls_conn_back_write (struct sock_mapping *map) +static bool pvcalls_conn_back_write (struct sock_mapping *map) {struct pvcalls_data_intf *intf = map->ring;struct pvcalls_data *data = &map->data;
@@ -189,7 +190,7 @@ static void pvcalls_conn_back_write(stru array_size =XEN_FLEX_RING_SIZE (map->ring_order); size =pvcalls_queued (prod, cons, array_size);if (size ==0 ) -return ; +return false;memset (&msg,0 ,sizeof (msg)); msg.msg_flags |=MSG_DONTWAIT ;
@@ -207,12 +208,11 @@ static void pvcalls_conn_back_write(struatomic_set (&map->write,0 ); ret =inet_sendmsg (map->sock, &msg, size); -if (ret == -EAGAIN || (ret >=0 && ret < size)) { +if (ret == -EAGAIN ) {atomic_inc (&map->write);atomic_inc (&map->io); +return true; } -if (ret == -EAGAIN ) -return ;/* write the data, then update the indexes */ virt_wmb ();
@@ -225,9 +225,13 @@ static void pvcalls_conn_back_write(stru }/* update the indexes, then notify the other end */ virt_wmb (); -if (prod != cons + ret) +if (prod != cons + ret) {atomic_inc (&map->write); +atomic_inc (&map->io); + }notify_remote_via_irq (map->irq); + +return true; }static void pvcalls_back_ioworker (struct work_struct *work)
@@ -236,6 +240,7 @@ static void pvcalls_back_ioworker(struct struct pvcalls_ioworker, register_work); struct sock_mapping *map = container_of(ioworker, struct sock_mapping, ioworker); +unsigned int eoi_flags =XEN_EOI_FLAG_SPURIOUS ;while (atomic_read (&map->io) >0 ) {if (atomic_read (&map->release) >0 ) {
@@ -243,10 +248,18 @@ static void pvcalls_back_ioworker(structreturn ; } -if (atomic_read (&map->read) >0 ) -pvcalls_conn_back_read (map); -if (atomic_read (&map->write) >0 ) -pvcalls_conn_back_write (map); +if (atomic_read (&map->read) >0 && +pvcalls_conn_back_read (map)) + eoi_flags =0 ; +if (atomic_read (&map->write) >0 && +pvcalls_conn_back_write (map)) + eoi_flags =0 ; + +if (atomic_read (&map->eoi) >0 && !atomic_read (&map->write)) { +atomic_set (&map->eoi,0 ); +xen_irq_lateeoi (map->irq, eoi_flags); + eoi_flags =XEN_EOI_FLAG_SPURIOUS ; + }atomic_dec (&map->io); }
@@ -343,12 +356,9 @@ static struct sock_mapping *pvcalls_new_ goto out; map->bytes = page; - ret =bind_interdomain_evtchn_to_irqhandler (fedata->dev->otherend_id, - evtchn, - pvcalls_back_conn_event, -0 , -"pvcalls-backend" , - map); + ret =bind_interdomain_evtchn_to_irqhandler_lateeoi ( + fedata->dev->otherend_id, evtchn, + pvcalls_back_conn_event,0 ,"pvcalls-backend" , map);if (ret <0 ) goto out; map->irq = ret;
@@ -882,15 +892,18 @@ static irqreturn_t pvcalls_back_event(in {struct xenbus_device *dev = dev_id;struct pvcalls_fedata *fedata =NULL ; +unsigned int eoi_flags =XEN_EOI_FLAG_SPURIOUS ; -if (dev ==NULL ) -return IRQ_HANDLED ; +if (dev) { + fedata =dev_get_drvdata (&dev->dev); +if (fedata) { +pvcalls_back_work (fedata); + eoi_flags =0 ; + } + } - fedata =dev_get_drvdata (&dev->dev); -if (fedata ==NULL ) -return IRQ_HANDLED ; +xen_irq_lateeoi (irq, eoi_flags); -pvcalls_back_work (fedata);return IRQ_HANDLED ; }
@@ -900,12 +913,15 @@ static irqreturn_t pvcalls_back_conn_evestruct pvcalls_ioworker *iow;if (map ==NULL || map->sock ==NULL || map->sock->sk ==NULL || - map->sock->sk->sk_user_data != map) + map->sock->sk->sk_user_data != map) { +xen_irq_lateeoi (irq,0 );return IRQ_HANDLED ; + } iow = &map->ioworker;atomic_inc (&map->write); +atomic_inc (&map->eoi);atomic_inc (&map->io);queue_work (iow->wq, &iow->register_work);
@@ -940,7 +956,7 @@ static int backend_connect(struct xenbus goto error; } - err =bind_interdomain_evtchn_to_irq (dev->otherend_id, evtchn); + err =bind_interdomain_evtchn_to_irq_lateeoi (dev->otherend_id, evtchn);if (err <0 ) goto error; fedata->irq = err;
Greg Kroah-Hartman
maintainerPATCH 45/191 xen/pciback: use lateeoi irq binding
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit c2711441bc961b37bba0615dd7135857d189035f upstream.
In order to reduce the chance for the system becoming unresponsive due to event storms triggered by a misbehaving pcifront use the lateeoi irq binding for pciback and unmask the event channel only just before leaving the event handling function.
Restructure the handling to support that scheme. Basically an event can come in for two reasons: either a normal request for a pciback action, which is handled in a worker, or in case the guest has finished an AER request which was requested by pciback.
When an AER request is issued to the guest and a normal pciback action is currently active issue an EOI early in order to be able to receive another event when the AER request has been finished by the guest.
Let the worker processing the normal requests run until no further request is pending, instead of starting a new worker ion that case. Issue the EOI only just before leaving the worker.
This scheme allows to drop calling the generic function xen_pcibk_test_and_schedule_op() after processing of any request as the handling of both request types is now separated more cleanly.
This is part of XSA-332.
--- drivers/xen/xen-pciback/pci_stub.c | 14 ++++----- drivers/xen/xen-pciback/pciback.h | 12 +++++++- drivers/xen/xen-pciback/pciback_ops.c | 48 ++++++++++++++++++++++++++-------- drivers/xen/xen-pciback/xenbus.c | 2 - 4 files changed, 56 insertions(+), 20 deletions(-)
drivers/xen/xen-pciback/pci_stub.c +7 -7
@@ -733,10 +733,17 @@ static pci_ers_result_t common_process(swmb ();notify_remote_via_irq (pdev->evtchn_irq); +/* Enable IRQ to signal "request done". */ +xen_pcibk_lateeoi (pdev,0 ); + ret =wait_event_timeout (xen_pcibk_aer_wait_queue, !(test_bit (_XEN_PCIB_active, (unsigned long *) &sh_info->flags)),300 *HZ ); +/* Enable IRQ for pcifront request if not already active. */ +if (!test_bit (_PDEVF_op_active, &pdev->flags)) +xen_pcibk_lateeoi (pdev,0 ); +if (!ret) {if (test_bit (_XEN_PCIB_active, (unsigned long *)&sh_info->flags)) {
@@ -750,13 +757,6 @@ static pci_ers_result_t common_process(s }clear_bit (_PCIB_op_pending, (unsigned long *)&pdev->flags); -if (test_bit (_XEN_PCIF_active, - (unsigned long *)&sh_info->flags)) { -dev_dbg (&psdev->dev->dev, -"schedule pci_conf service in " DRV_NAME "\n" ); -xen_pcibk_test_and_schedule_op (psdev->pdev); - } - res = (pci_ers_result_t )aer_op->err;return res; }
drivers/xen/xen-pciback/pciback.h +10 -2
@@ -14,6 +14,7 @@#include <linux/spinlock.h> #include <linux/workqueue.h> #include <linux/atomic.h> +#include <xen/events.h> #include <xen/interface/io/pciif.h> #define DRV_NAME "xen-pciback"
@@ -27,6 +28,8 @@ struct pci_dev_entry {#define PDEVF_op_active (1<<(_PDEVF_op_active))#define _PCIB_op_pending (1)#define PCIB_op_pending (1<<(_PCIB_op_pending)) +#define _EOI_pending (2) +#define EOI_pending (1<<(_EOI_pending))struct xen_pcibk_device {void *pci_dev_data;
@@ -182,12 +185,17 @@ static inline void xen_pcibk_release_devirqreturn_t xen_pcibk_handle_event (int irq,void *dev_id);void xen_pcibk_do_op (struct work_struct *data); +static inline void xen_pcibk_lateeoi (struct xen_pcibk_device *pdev, +unsigned int eoi_flag) +{ +if (test_and_clear_bit (_EOI_pending, &pdev->flags)) +xen_irq_lateeoi (pdev->evtchn_irq, eoi_flag); +} +int xen_pcibk_xenbus_register (void );void xen_pcibk_xenbus_unregister (void );extern int verbose_request; - -void xen_pcibk_test_and_schedule_op (struct xen_pcibk_device *pdev);#endif /* Handles shared IRQs that can to device domain and control domain. */
drivers/xen/xen-pciback/pciback_ops.c +38 -10
90 lines not on this page · raw message
drivers/xen/xen-pciback/xenbus.c +1 -1
@@ -123,7 +123,7 @@ static int xen_pcibk_do_attach(struct xe pdev->sh_info = vaddr; - err =bind_interdomain_evtchn_to_irqhandler ( + err =bind_interdomain_evtchn_to_irqhandler_lateeoi ( pdev->xdev->otherend_id, remote_evtchn, xen_pcibk_handle_event,0 ,DRV_NAME , pdev);if (err <0 ) {
Greg Kroah-Hartman
maintainerPATCH 46/191 xen/events: switch user event channels to lateeoi model
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit c44b849cee8c3ac587da3b0980e01f77500d158c upstream.
Instead of disabling the irq when an event is received and enabling it again when handled by the user process use the lateeoi model.
This is part of XSA-332.
--- drivers/xen/evtchn.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/xen/evtchn.c +3 -4
@@ -166,7 +166,6 @@ static irqreturn_t evtchn_interrupt(int"Interrupt for port %d, but apparently not enabled; per-user %p\n" , evtchn->port, u); -disable_irq_nosync (irq); evtchn->enabled = false;spin_lock (&u->ring_prod_lock);
@@ -292,7 +291,7 @@ static ssize_t evtchn_write(struct file evtchn =find_evtchn (u, port);if (evtchn && !evtchn->enabled) { evtchn->enabled = true; -enable_irq (irq_from_evtchn (port)); +xen_irq_lateeoi (irq_from_evtchn (port),0 ); } }
@@ -392,8 +391,8 @@ static int evtchn_bind_to_user(struct peif (rc <0 ) goto err; - rc =bind_evtchn_to_irqhandler (port, evtchn_interrupt,0 , - u->name, evtchn); + rc =bind_evtchn_to_irqhandler_lateeoi (port, evtchn_interrupt,0 , + u->name, evtchn);if (rc <0 ) goto err;
Greg Kroah-Hartman
maintainerPATCH 47/191 xen/events: use a common cpu hotplug hook for event channels
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 7beb290caa2adb0a399e735a1e175db9aae0523a upstream.
Today only fifo event channels have a cpu hotplug callback. In order to prepare for more percpu (de)init work move that callback into events_base.c and add percpu_init() and percpu_deinit() hooks to struct evtchn_ops.
This is part of XSA-332.
--- drivers/xen/events/events_base.c | 25 +++++++++++++++++++++ drivers/xen/events/events_fifo.c | 40 ++++++++++++++++------------------- drivers/xen/events/events_internal.h | 3 ++ 3 files changed, 47 insertions(+), 21 deletions(-)
drivers/xen/events/events_base.c +25 -0
@@ -33,6 +33,7 @@#include <linux/irqnr.h> #include <linux/pci.h> #include <linux/spinlock.h> +#include <linux/cpuhotplug.h> #ifdef CONFIG_X86 #include <asm/desc.h>
@@ -1833,6 +1834,26 @@ void xen_callback_vector(void) {}static bool fifo_events = true;module_param (fifo_events, bool,0 ); +static int xen_evtchn_cpu_prepare (unsigned int cpu) +{ +int ret =0 ; + +if (evtchn_ops->percpu_init) + ret = evtchn_ops->percpu_init (cpu); + +return ret; +} + +static int xen_evtchn_cpu_dead (unsigned int cpu) +{ +int ret =0 ; + +if (evtchn_ops->percpu_deinit) + ret = evtchn_ops->percpu_deinit (cpu); + +return ret; +} +void __init xen_init_IRQ (void ) {int ret = -EINVAL ;
@@ -1843,6 +1864,10 @@ void __init xen_init_IRQ(void)if (ret <0 )xen_evtchn_2l_init (); +cpuhp_setup_state_nocalls (CPUHP_XEN_EVTCHN_PREPARE , +"xen/evtchn:prepare" , + xen_evtchn_cpu_prepare, xen_evtchn_cpu_dead); + evtchn_to_irq =kcalloc (EVTCHN_ROW (xen_evtchn_max_channels ()),sizeof (*evtchn_to_irq),GFP_KERNEL );BUG_ON (!evtchn_to_irq);
drivers/xen/events/events_fifo.c +19 -21
@@ -385,21 +385,6 @@ static void evtchn_fifo_resume(void) event_array_pages =0 ; } -static const struct evtchn_ops evtchn_ops_fifo = { - .max_channels = evtchn_fifo_max_channels, - .nr_channels = evtchn_fifo_nr_channels, - .setup = evtchn_fifo_setup, - .bind_to_cpu = evtchn_fifo_bind_to_cpu, - .clear_pending = evtchn_fifo_clear_pending, - .set_pending = evtchn_fifo_set_pending, - .is_pending = evtchn_fifo_is_pending, - .test_and_set_mask = evtchn_fifo_test_and_set_mask, - .mask = evtchn_fifo_mask, - .unmask = evtchn_fifo_unmask, - .handle_events = evtchn_fifo_handle_events, - .resume = evtchn_fifo_resume, -}; -static int evtchn_fifo_alloc_control_block (unsigned cpu) {void *control_block =NULL ;
@@ -422,19 +407,36 @@ static int evtchn_fifo_alloc_control_bloreturn ret; } -static int xen_evtchn_cpu_prepare (unsigned int cpu) +static int evtchn_fifo_percpu_init (unsigned int cpu) {if (!per_cpu (cpu_control_block, cpu))return evtchn_fifo_alloc_control_block (cpu);return 0 ; } -static int xen_evtchn_cpu_dead (unsigned int cpu) +static int evtchn_fifo_percpu_deinit (unsigned int cpu) {__evtchn_fifo_handle_events (cpu, true);return 0 ; } +static const struct evtchn_ops evtchn_ops_fifo = { + .max_channels = evtchn_fifo_max_channels, + .nr_channels = evtchn_fifo_nr_channels, + .setup = evtchn_fifo_setup, + .bind_to_cpu = evtchn_fifo_bind_to_cpu, + .clear_pending = evtchn_fifo_clear_pending, + .set_pending = evtchn_fifo_set_pending, + .is_pending = evtchn_fifo_is_pending, + .test_and_set_mask = evtchn_fifo_test_and_set_mask, + .mask = evtchn_fifo_mask, + .unmask = evtchn_fifo_unmask, + .handle_events = evtchn_fifo_handle_events, + .resume = evtchn_fifo_resume, + .percpu_init = evtchn_fifo_percpu_init, + .percpu_deinit = evtchn_fifo_percpu_deinit, +}; +int __init xen_evtchn_fifo_init (void ) {int cpu =smp_processor_id ();
@@ -448,9 +450,5 @@ int __init xen_evtchn_fifo_init(void) evtchn_ops = &evtchn_ops_fifo; -cpuhp_setup_state_nocalls (CPUHP_XEN_EVTCHN_PREPARE , -"xen/evtchn:prepare" , - xen_evtchn_cpu_prepare, xen_evtchn_cpu_dead); -return ret; }
drivers/xen/events/events_internal.h +3 -0
@@ -71,6 +71,9 @@ struct evtchn_ops {void (*handle_events)(unsigned cpu);void (*resume)(void ); + +int (*percpu_init)(unsigned int cpu); +int (*percpu_deinit)(unsigned int cpu); };extern const struct evtchn_ops *evtchn_ops;
Greg Kroah-Hartman
maintainerPATCH 48/191 xen/events: defer eoi in case of excessive number of events
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit e99502f76271d6bc4e374fe368c50c67a1fd3070 upstream.
In case rogue guests are sending events at high frequency it might happen that xen_evtchn_do_upcall() won't stop processing events in dom0. As this is done in irq handling a crash might be the result.
In order to avoid that, delay further inter-domain events after some time in xen_evtchn_do_upcall() by forcing eoi processing into a worker on the same cpu, thus inhibiting new events coming in.
The time after which eoi processing is to be delayed is configurable via a new module parameter "event_loop_timeout" which specifies the maximum event loop time in jiffies (default: 2, the value was chosen after some tests showing that a value of 2 was the lowest with an only slight drop of dom0 network throughput while multiple guests performed an event storm).
How long eoi processing will be delayed can be specified via another parameter "event_eoi_delay" (again in jiffies, default 10, again the value was chosen after testing with different delay values).
This is part of XSA-332.
--- Documentation/admin-guide/kernel-parameters.txt | 8 + drivers/xen/events/events_2l.c | 7 drivers/xen/events/events_base.c | 189 +++++++++++++++++++++++- drivers/xen/events/events_fifo.c | 30 +-- drivers/xen/events/events_internal.h | 14 + 5 files changed, 216 insertions(+), 32 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +8 -0
@@ -5270,6 +5270,14 @@ with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. + xen.event_eoi_delay= [XEN] + How long to delay EOI handling in case of event + storms (jiffies). Default is 10. + + xen.event_loop_timeout= [XEN] + After which time (jiffies) the event handling loop + should start to delay EOI handling. Default is 2. + xirc2ps_cs= [NET,PCMCIA] Format: <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
drivers/xen/events/events_2l.c +2 -5
@@ -161,7 +161,7 @@ static inline xen_ulong_t active_evtchns* a bitset of words which contain pending event bits. The second * level is a bitset of pending events themselves. */ -static void evtchn_2l_handle_events (unsigned cpu) +static void evtchn_2l_handle_events (unsigned cpu,struct evtchn_loop_ctrl *ctrl) {int irq;xen_ulong_t pending_words;
@@ -242,10 +242,7 @@ static void evtchn_2l_handle_events(unsi/* Process port. */ port = (word_idx *BITS_PER_EVTCHN_WORD ) + bit_idx; - irq =get_evtchn_to_irq (port); - -if (irq !=-1 ) -generic_handle_irq (irq); +handle_irq_for_port (port, ctrl); bit_idx = (bit_idx +1 ) %BITS_PER_EVTCHN_WORD ;
drivers/xen/events/events_base.c +184 -5
289 lines not on this page · raw message
drivers/xen/events/events_fifo.c +11 -19
@@ -275,19 +275,9 @@ static uint32_t clear_linked(volatile evreturn w &EVTCHN_FIFO_LINK_MASK ; } -static void handle_irq_for_port (unsigned port) -{ -int irq; - - irq =get_evtchn_to_irq (port); -if (irq !=-1 ) -generic_handle_irq (irq); -} - -static void consume_one_event (unsigned cpu, +static void consume_one_event (unsigned cpu,struct evtchn_loop_ctrl *ctrl,struct evtchn_fifo_control_block *control_block, -unsigned priority,unsigned long *ready, -bool drop) +unsigned priority,unsigned long *ready) {struct evtchn_fifo_queue *q = &per_cpu (cpu_queue, cpu);uint32_t head;
@@ -320,16 +310,17 @@ static void consume_one_event(unsigned cclear_bit (priority, ready);if (evtchn_fifo_is_pending (port) && !evtchn_fifo_is_masked (port)) { -if (unlikely (drop)) +if (unlikely (!ctrl))pr_warn ("Dropping pending event for port %u\n" , port);else -handle_irq_for_port (port); +handle_irq_for_port (port, ctrl); } q->head[priority] = head; } -static void __evtchn_fifo_handle_events (unsigned cpu,bool drop) +static void __evtchn_fifo_handle_events (unsigned cpu, +struct evtchn_loop_ctrl *ctrl) {struct evtchn_fifo_control_block *control_block;unsigned long ready;
@@ -341,14 +332,15 @@ static void __evtchn_fifo_handle_events(while (ready) { q =find_first_bit (&ready,EVTCHN_FIFO_MAX_QUEUES ); -consume_one_event (cpu, control_block, q, &ready, drop); +consume_one_event (cpu, ctrl, control_block, q, &ready); ready |=xchg (&control_block->ready,0 ); } } -static void evtchn_fifo_handle_events (unsigned cpu) +static void evtchn_fifo_handle_events (unsigned cpu, +struct evtchn_loop_ctrl *ctrl) { -__evtchn_fifo_handle_events (cpu, false); +__evtchn_fifo_handle_events (cpu, ctrl); }static void evtchn_fifo_resume (void )
@@ -416,7 +408,7 @@ static int evtchn_fifo_percpu_init(unsigstatic int evtchn_fifo_percpu_deinit (unsigned int cpu) { -__evtchn_fifo_handle_events (cpu, true); +__evtchn_fifo_handle_events (cpu,NULL );return 0 ; }
drivers/xen/events/events_internal.h +11 -3
@@ -32,11 +32,15 @@ enum xen_irq_type {*/ struct irq_info {struct list_head list; +struct list_head eoi_list;int refcnt;enum xen_irq_type type;/* type */ unsigned irq;unsigned int evtchn;/* event channel */ unsigned short cpu;/* cpu bound */ +unsigned short eoi_cpu;/* EOI must happen on this cpu */ +unsigned int irq_epoch;/* If eoi_cpu valid: irq_epoch of event */ +u64 eoi_time;/* Time in jiffies when to EOI. */ union {unsigned short virq;
@@ -55,6 +59,8 @@ struct irq_info {#define PIRQ_SHAREABLE (1 << 1)#define PIRQ_MSI_GROUP (1 << 2) +struct evtchn_loop_ctrl ; +struct evtchn_ops {unsigned (*max_channels)(void );unsigned (*nr_channels)(void );
@@ -69,7 +75,7 @@ struct evtchn_ops {void (*mask)(unsigned port);void (*unmask)(unsigned port); -void (*handle_events)(unsigned cpu); +void (*handle_events)(unsigned cpu,struct evtchn_loop_ctrl *ctrl);void (*resume)(void );int (*percpu_init)(unsigned int cpu);
@@ -80,6 +86,7 @@ extern const struct evtchn_ops *evtchn_oextern int **evtchn_to_irq;int get_evtchn_to_irq (unsigned int evtchn); +void handle_irq_for_port (evtchn_port_t port,struct evtchn_loop_ctrl *ctrl);struct irq_info *info_for_irq (unsigned irq);unsigned cpu_from_irq (unsigned irq);
@@ -137,9 +144,10 @@ static inline void unmask_evtchn(unsignereturn evtchn_ops->unmask (port); } -static inline void xen_evtchn_handle_events (unsigned cpu) +static inline void xen_evtchn_handle_events (unsigned cpu, +struct evtchn_loop_ctrl *ctrl) { -return evtchn_ops->handle_events (cpu); +return evtchn_ops->handle_events (cpu, ctrl); }static inline void xen_evtchn_resume (void )
Greg Kroah-Hartman
maintainerPATCH 49/191 xen/events: block rogue events for some time
In reply to Greg Kroah-HartmanFrom: Juergen Gross <[email protected]>
commit 5f7f77400ab5b357b5fdb7122c3442239672186c upstream.
In order to avoid high dom0 load due to rogue guests sending events at high frequency, block those events in case there was no action needed in dom0 to handle the events.
This is done by adding a per-event counter, which set to zero in case an EOI without the XEN_EOI_FLAG_SPURIOUS is received from a backend driver, and incremented when this flag has been set. In case the counter is 2 or higher delay the EOI by 1 << (cnt - 2) jiffies, but not more than 1 second.
In order not to waste memory shorten the per-event refcnt to two bytes (it should normally never exceed a value of 2). Add an overflow check to evtchn_get() to make sure the 2 bytes really won't overflow.
This is part of XSA-332.
--- drivers/xen/events/events_base.c | 27 ++++++++++++++++++++++----- drivers/xen/events/events_internal.h | 3 ++- 2 files changed, 24 insertions(+), 6 deletions(-)
drivers/xen/events/events_base.c +22 -5
@@ -459,17 +459,34 @@ static void lateeoi_list_add(struct irq_spin_unlock_irqrestore (&eoi->eoi_list_lock, flags); } -static void xen_irq_lateeoi_locked (struct irq_info *info) +static void xen_irq_lateeoi_locked (struct irq_info *info,bool spurious) {evtchn_port_t evtchn;unsigned int cpu; +unsigned int delay =0 ; evtchn = info->evtchn;if (!VALID_EVTCHN (evtchn) || !list_empty (&info->eoi_list))return ; +if (spurious) { +if ((1 << info->spurious_cnt) < (HZ <<2 )) + info->spurious_cnt++; +if (info->spurious_cnt >1 ) { + delay =1 << (info->spurious_cnt -2 ); +if (delay >HZ ) + delay =HZ ; +if (!info->eoi_time) + info->eoi_cpu =smp_processor_id (); + info->eoi_time =get_jiffies_64 () + delay; + } + }else { + info->spurious_cnt =0 ; + } + cpu = info->eoi_cpu; -if (info->eoi_time && info->irq_epoch ==per_cpu (irq_epoch, cpu)) { +if (info->eoi_time && + (info->irq_epoch ==per_cpu (irq_epoch, cpu) || delay)) {lateeoi_list_add (info);return ; }
@@ -506,7 +523,7 @@ static void xen_irq_lateeoi_worker(struc info->eoi_time =0 ; -xen_irq_lateeoi_locked (info); +xen_irq_lateeoi_locked (info, false); }if (info)
@@ -535,7 +552,7 @@ void xen_irq_lateeoi(unsigned int irq, u info =info_for_irq (irq);if (info) -xen_irq_lateeoi_locked (info); +xen_irq_lateeoi_locked (info, eoi_flags &XEN_EOI_FLAG_SPURIOUS );read_unlock_irqrestore (&evtchn_rwlock, flags); }
@@ -1438,7 +1455,7 @@ int evtchn_get(unsigned int evtchn) goto done; err = -EINVAL ; -if (info->refcnt <=0 ) +if (info->refcnt <=0 || info->refcnt ==SHRT_MAX ) goto done; info->refcnt++;
drivers/xen/events/events_internal.h +2 -1
@@ -33,7 +33,8 @@ enum xen_irq_type {struct irq_info {struct list_head list;struct list_head eoi_list; -int refcnt; +short refcnt; +short spurious_cnt;enum xen_irq_type type;/* type */ unsigned irq;unsigned int evtchn;/* event channel */
Greg Kroah-Hartman
maintainerPATCH 50/191 x86/unwind/orc: Fix inactive tasks with stack pointer in %sp on GCC 10 compiled kernels
In reply to Greg Kroah-HartmanFrom: Jiri Slaby <[email protected]>
[ Upstream commit f2ac57a4c49d40409c21c82d23b5706df9b438af ]
GCC 10 optimizes the scheduler code differently than its predecessors.
When CONFIG_DEBUG_SECTION_MISMATCH=y, the Makefile forces GCC not to inline some functions (-fno-inline-functions-called-once). Before GCC 10, "no-inlined" __schedule() starts with the usual prologue:
push %bp
mov %sp, %bp
So the ORC unwinder simply picks stack pointer from %bp and unwinds from __schedule() just perfectly:
$ cat /proc/1/stack [<0>] ep_poll+0x3e9/0x450 [<0>] do_epoll_wait+0xaa/0xc0 [<0>] __x64_sys_epoll_wait+0x1a/0x20 [<0>] do_syscall_64+0x33/0x40 [<0>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
But now, with GCC 10, there is no %bp prologue in __schedule():
$ cat /proc/1/stack <nothing>
The ORC entry of the point in __schedule() is:
sp:sp+88 bp:last_sp-48 type:call end:0
In this case, nobody subtracts sizeof "struct inactive_task_frame" in __unwind_start(). The struct is put on the stack by __switch_to_asm() and only then __switch_to_asm() stores %sp to task->thread.sp. But we start unwinding from a point in __schedule() (stored in frame->ret_addr by 'call') and not in __switch_to_asm().
So for these example values in __unwind_start():
sp=ffff94b50001fdc8 bp=ffff8e1f41d29340 ip=__schedule+0x1f0
The stack is:
ffff94b50001fdc8: ffff8e1f41578000 # struct inactive_task_frame
ffff94b50001fdd0: 0000000000000000
ffff94b50001fdd8: ffff8e1f41d29340
ffff94b50001fde0: ffff8e1f41611d40 # ...
ffff94b50001fde8: ffffffff93c41920 # bx
ffff94b50001fdf0: ffff8e1f41d29340 # bp
ffff94b50001fdf8: ffffffff9376cad0 # ret_addr (and end of the struct)
0xffffffff9376cad0 is __schedule+0x1f0 (after the call to __switch_to_asm). Now follow those 88 bytes from the ORC entry (sp+88). The entry is correct, __schedule() really pushes 48 bytes (8*7) + 32 bytes via subq to store some local values (like 4U below). So to unwind, look at the offset 88-sizeof(long) = 0x50 from here:
ffff94b50001fe00: ffff8e1f41578618
ffff94b50001fe08: 00000cc000000255
ffff94b50001fe10: 0000000500000004
ffff94b50001fe18: 7793fab6956b2d00 # NOTE (see below)
ffff94b50001fe20: ffff8e1f41578000
ffff94b50001fe28: ffff8e1f41578000
ffff94b50001fe30: ffff8e1f41578000
ffff94b50001fe38: ffff8e1f41578000
ffff94b50001fe40: ffff94b50001fed8
ffff94b50001fe48: ffff8e1f41577ff0
ffff94b50001fe50: ffffffff9376cf12
Here ^^^^^^^^^^^^^^^^ is the correct ret addr from__schedule(). It translates to schedule+0x42 (insn after a call to __schedule()).
BUT, unwind_next_frame() tries to take the address starting from 0xffff94b50001fdc8. That is exactly from thread.sp+88-sizeof(long) = 0xffff94b50001fdc8+88-8 = 0xffff94b50001fe18, which is garbage marked as NOTE above. So this quits the unwinding as 7793fab6956b2d00 is obviously not a kernel address.
There was a fix to skip 'struct inactive_task_frame' in unwind_get_return_address_ptr in the following commit:
187b96db5ca7 ("x86/unwind/orc: Fix unwind_get_return_address_ptr() for inactive tasks")
But we need to skip the struct already in the unwinder proper. So subtract the size (increase the stack pointer) of the structure in __unwind_start() directly. This allows for removal of the code added by commit 187b96db5ca7 completely, as the address is now at '(unsigned long *)state->sp - 1', the same as in the generic case.
[ mingo: Cleaned up the changelog a bit, for better readability. ]
--- arch/x86/kernel/unwind_orc.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-)
arch/x86/kernel/unwind_orc.c +1 -8
@@ -300,19 +300,12 @@ EXPORT_SYMBOL_GPL(unwind_get_return_address);unsigned long *unwind_get_return_address_ptr (struct unwind_state *state) { -struct task_struct *task = state->task; -if (unwind_done (state))return NULL ;if (state->regs)return &state->regs->ip; -if (task != current && state->sp == task->thread.sp) { -struct inactive_task_frame *frame = (void *)task->thread.sp; -return &frame->ret_addr; - } -if (state->sp)return (unsigned long *)state->sp -1 ;
@@ -634,7 +627,7 @@ void __unwind_start(struct unwind_state *state, struct task_struct *task, }else {struct inactive_task_frame *frame = (void *)task->thread.sp; - state->sp = task->thread.sp; + state->sp = task->thread.sp +sizeof (*frame); state->bp =READ_ONCE_NOCHECK (frame->bp); state->ip =READ_ONCE_NOCHECK (frame->ret_addr); state->signal = (void *)state->ip == ret_from_fork;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 51/191 mlxsw: core: Fix use-after-free in mlxsw_emad_trans_finish()
In reply to Greg Kroah-HartmanFrom: Amit Cohen <[email protected]>
[ Upstream commit 0daf2bf5a2dcf33d446b76360908f109816e2e21 ]
Each EMAD transaction stores the skb used to issue the EMAD request ('trans->tx_skb') so that the request could be retried in case of a timeout. The skb can be freed when a corresponding response is received or as part of the retry logic (e.g., failed retransmit, exceeded maximum number of retries).
The two tasks (i.e., response processing and retransmits) are synchronized by the atomic 'trans->active' field which ensures that responses to inactive transactions are ignored.
In case of a failed retransmit the transaction is finished and all of its resources are freed. However, the current code does not mark it as inactive. Syzkaller was able to hit a race condition in which a concurrent response is processed while the transaction's resources are being freed, resulting in a use-after-free [1].
Fix the issue by making sure to mark the transaction as inactive after a failed retransmit and free its resources only if a concurrent task did not already do that.
[1]
BUG: KASAN: use-after-free in consume_skb+0x30/0x370
net/core/skbuff.c:833
Read of size 4 at addr ffff88804f570494 by task syz-executor.0/1004
CPU: 0 PID: 1004 Comm: syz-executor.0 Not tainted 5.8.0-rc7+ #68
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0xf6/0x16e lib/dump_stack.c:118
print_address_description.constprop.0+0x1c/0x250
mm/kasan/report.c:383
__kasan_report mm/kasan/report.c:513 [inline]
kasan_report.cold+0x1f/0x37 mm/kasan/report.c:530
check_memory_region_inline mm/kasan/generic.c:186 [inline]
check_memory_region+0x14e/0x1b0 mm/kasan/generic.c:192
instrument_atomic_read include/linux/instrumented.h:56 [inline]
atomic_read include/asm-generic/atomic-instrumented.h:27 [inline]
refcount_read include/linux/refcount.h:147 [inline]
skb_unref include/linux/skbuff.h:1044 [inline]
consume_skb+0x30/0x370 net/core/skbuff.c:833
mlxsw_emad_trans_finish+0x64/0x1c0 drivers/net/ethernet/mellanox/mlxsw/core.c:592
mlxsw_emad_process_response drivers/net/ethernet/mellanox/mlxsw/core.c:651 [inline]
mlxsw_emad_rx_listener_func+0x5c9/0xac0 drivers/net/ethernet/mellanox/mlxsw/core.c:672
mlxsw_core_skb_receive+0x4df/0x770 drivers/net/ethernet/mellanox/mlxsw/core.c:2063
mlxsw_pci_cqe_rdq_handle drivers/net/ethernet/mellanox/mlxsw/pci.c:595 [inline]
mlxsw_pci_cq_tasklet+0x12a6/0x2520 drivers/net/ethernet/mellanox/mlxsw/pci.c:651
tasklet_action_common.isra.0+0x13f/0x3e0 kernel/softirq.c:550
__do_softirq+0x223/0x964 kernel/softirq.c:292
asm_call_on_stack+0x12/0x20 arch/x86/entry/entry_64.S:711
Allocated by task 1006:
save_stack+0x1b/0x40 mm/kasan/common.c:48
set_track mm/kasan/common.c:56 [inline]
__kasan_kmalloc mm/kasan/common.c:494 [inline]
__kasan_kmalloc.constprop.0+0xc2/0xd0 mm/kasan/common.c:467
slab_post_alloc_hook mm/slab.h:586 [inline]
slab_alloc_node mm/slub.c:2824 [inline]
slab_alloc mm/slub.c:2832 [inline]
kmem_cache_alloc+0xcd/0x2e0 mm/slub.c:2837
__build_skb+0x21/0x60 net/core/skbuff.c:311
__netdev_alloc_skb+0x1e2/0x360 net/core/skbuff.c:464
netdev_alloc_skb include/linux/skbuff.h:2810 [inline]
mlxsw_emad_alloc drivers/net/ethernet/mellanox/mlxsw/core.c:756 [inline]
mlxsw_emad_reg_access drivers/net/ethernet/mellanox/mlxsw/core.c:787 [inline]
mlxsw_core_reg_access_emad+0x1ab/0x1420 drivers/net/ethernet/mellanox/mlxsw/core.c:1817
mlxsw_reg_trans_query+0x39/0x50 drivers/net/ethernet/mellanox/mlxsw/core.c:1831
mlxsw_sp_sb_pm_occ_clear drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c:260 [inline]
mlxsw_sp_sb_occ_max_clear+0xbff/0x10a0 drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c:1365
mlxsw_devlink_sb_occ_max_clear+0x76/0xb0 drivers/net/ethernet/mellanox/mlxsw/core.c:1037
devlink_nl_cmd_sb_occ_max_clear_doit+0x1ec/0x280 net/core/devlink.c:1765
genl_family_rcv_msg_doit net/netlink/genetlink.c:669 [inline]
genl_family_rcv_msg net/netlink/genetlink.c:714 [inline]
genl_rcv_msg+0x617/0x980 net/netlink/genetlink.c:731
netlink_rcv_skb+0x152/0x440 net/netlink/af_netlink.c:2470
genl_rcv+0x24/0x40 net/netlink/genetlink.c:742
netlink_unicast_kernel net/netlink/af_netlink.c:1304 [inline]
netlink_unicast+0x53a/0x750 net/netlink/af_netlink.c:1330
netlink_sendmsg+0x850/0xd90 net/netlink/af_netlink.c:1919
sock_sendmsg_nosec net/socket.c:651 [inline]
sock_sendmsg+0x150/0x190 net/socket.c:671
____sys_sendmsg+0x6d8/0x840 net/socket.c:2359
___sys_sendmsg+0xff/0x170 net/socket.c:2413
__sys_sendmsg+0xe5/0x1b0 net/socket.c:2446
do_syscall_64+0x56/0xa0 arch/x86/entry/common.c:384
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Freed by task 73:
save_stack+0x1b/0x40 mm/kasan/common.c:48
set_track mm/kasan/common.c:56 [inline]
kasan_set_free_info mm/kasan/common.c:316 [inline]
__kasan_slab_free+0x12c/0x170 mm/kasan/common.c:455
slab_free_hook mm/slub.c:1474 [inline]
slab_free_freelist_hook mm/slub.c:1507 [inline]
slab_free mm/slub.c:3072 [inline]
kmem_cache_free+0xbe/0x380 mm/slub.c:3088
kfree_skbmem net/core/skbuff.c:622 [inline]
kfree_skbmem+0xef/0x1b0 net/core/skbuff.c:616
__kfree_skb net/core/skbuff.c:679 [inline]
consume_skb net/core/skbuff.c:837 [inline]
consume_skb+0xe1/0x370 net/core/skbuff.c:831
mlxsw_emad_trans_finish+0x64/0x1c0 drivers/net/ethernet/mellanox/mlxsw/core.c:592
mlxsw_emad_transmit_retry.isra.0+0x9d/0xc0 drivers/net/ethernet/mellanox/mlxsw/core.c:613
mlxsw_emad_trans_timeout_work+0x43/0x50 drivers/net/ethernet/mellanox/mlxsw/core.c:625
process_one_work+0xa3e/0x17a0 kernel/workqueue.c:2269
worker_thread+0x9e/0x1050 kernel/workqueue.c:2415
kthread+0x355/0x470 kernel/kthread.c:291
ret_from_fork+0x22/0x30 arch/x86/entry/entry_64.S:293
The buggy address belongs to the object at ffff88804f5703c0
which belongs to the cache skbuff_head_cache of size 224
The buggy address is located 212 bytes inside of
224-byte region [ffff88804f5703c0, ffff88804f5704a0)
The buggy address belongs to the page:
page:ffffea00013d5c00 refcount:1 mapcount:0 mapping:0000000000000000
index:0x0
flags: 0x100000000000200(slab)
raw: 0100000000000200 dead000000000100 dead000000000122 ffff88806c625400
raw: 0000000000000000 00000000000c000c 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff88804f570380: fc fc fc fc fc fc fc fc fb fb fb fb fb fb fb fb
ffff88804f570400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88804f570480: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc
^
ffff88804f570500: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ffff88804f570580: 00 00 00 00 00 00 00 00 00 00 00 00 fc fc fc fc--- drivers/net/ethernet/mellanox/mlxsw/core.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/mellanox/mlxsw/core.c +3 -0
@@ -488,6 +488,9 @@ static void mlxsw_emad_transmit_retry(struct mlxsw_core *mlxsw_core, err =mlxsw_emad_transmit (trans->core, trans);if (err ==0 )return ; + +if (!atomic_dec_and_test (&trans->active)) +return ; }else { err = -EIO ; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 52/191 RDMA/qedr: Fix memory leak in iWARP CM
In reply to Greg Kroah-HartmanFrom: Alok Prasad <[email protected]>
[ Upstream commit a2267f8a52eea9096861affd463f691be0f0e8c9 ]
Fixes memory leak in iWARP CM
--- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 1 + 1 file changed, 1 insertion(+)
drivers/infiniband/hw/qedr/qedr_iw_cm.c +1 -0
@@ -677,6 +677,7 @@ int qedr_iw_destroy_listen(struct iw_cm_id *cm_id) listener->qed_handle); cm_id->rem_ref (cm_id); +kfree (listener);return rc; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 53/191 ata: sata_nv: Fix retrieving of active qcs
In reply to Greg Kroah-HartmanFrom: Sascha Hauer <[email protected]>
[ Upstream commit 8e4c309f9f33b76c09daa02b796ef87918eee494 ]
ata_qc_complete_multiple() has to be called with the tags physically active, that is the hw tag is at bit 0. ap->qc_active has the same tag at bit ATA_TAG_INTERNAL instead, so call ata_qc_get_active() to fix that up. This is done in the vein of 8385d756e114 ("libata: Fix retrieving of active qcs").
--- drivers/ata/sata_nv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ata/sata_nv.c +1 -1
@@ -2122,7 +2122,7 @@ static int nv_swncq_sdbfis(struct ata_port *ap) pp->dhfis_bits &= ~done_mask; pp->dmafis_bits &= ~done_mask; pp->sdbfis_bits |= done_mask; -ata_qc_complete_multiple (ap, ap->qc_active ^ done_mask); +ata_qc_complete_multiple (ap,ata_qc_get_active (ap) ^ done_mask);if (!ap->qc_active) {DPRINTK ("over\n" );
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 54/191 futex: Fix incorrect should_fail_futex() handling
In reply to Greg Kroah-HartmanFrom: Mateusz Nosek <[email protected]>
[ Upstream commit 921c7ebd1337d1a46783d7e15a850e12aed2eaa0 ]
If should_futex_fail() returns true in futex_wake_pi(), then the 'ret' variable is set to -EFAULT and then immediately overwritten. So the failure injection is non-functional.
Fix it by actually leaving the function and returning -EFAULT.
The Fixes tag is kinda blury because the initial commit which introduced failure injection was already sloppy, but the below mentioned commit broke it completely.
[ tglx: Massaged changelog ]
--- kernel/futex.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
kernel/futex.c +3 -1
@@ -1517,8 +1517,10 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_*/ newval =FUTEX_WAITERS |task_pid_vnr (new_owner); -if (unlikely (should_fail_futex (true))) +if (unlikely (should_fail_futex (true))) { ret = -EFAULT ; + goto out_unlock; + } ret =cmpxchg_futex_value_locked (&curval, uaddr, uval, newval);if (!ret && (curval != uval)) {
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 55/191 powerpc/powernv/smp: Fix spurious DBG() warning
In reply to Greg Kroah-HartmanFrom: Oliver O'Halloran <[email protected]>
[ Upstream commit f6bac19cf65c5be21d14a0c9684c8f560f2096dd ]
When building with W=1 we get the following warning:
arch/powerpc/platforms/powernv/smp.c: In function ‘pnv_smp_cpu_kill_self’:
arch/powerpc/platforms/powernv/smp.c:276:16: error: suggest braces around
empty body in an ‘if’ statement [-Werror=empty-body]
276 | cpu, srr1);
| ^
cc1: all warnings being treated as errors
The full context is this block:
if (srr1 && !generic_check_cpu_restart(cpu))
DBG("CPU%d Unexpected exit while offline srr1=%lx!\n",
cpu, srr1);
When building with DEBUG undefined DBG() expands to nothing and GCC emits the warning due to the lack of braces around an empty statement.
--- arch/powerpc/platforms/powernv/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/smp.c +1 -1
@@ -47,7 +47,7 @@#include <asm/udbg.h> #define DBG(fmt...) udbg_printf(fmt)#else -#define DBG(fmt...) +#define DBG(fmt...) do { } while (0)#endif static void pnv_smp_setup_cpu (int cpu)
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 56/191 powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
In reply to Greg Kroah-HartmanFrom: Nicholas Piggin <[email protected]>
[ Upstream commit 66acd46080bd9e5ad2be4b0eb1d498d5145d058e ]
powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing ARCH_WANT_IRQS_OFF_ACTIVATE_MM.
--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)
arch/powerpc/Kconfig +1 -0
@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MM select ARCH_WEAK_RELEASE_ACQUIRE select BINFMT_ELF select BUILDTIME_EXTABLE_SORT
arch/powerpc/include/asm/mmu_context.h +1 -1
@@ -204,7 +204,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,*/ static inline void activate_mm (struct mm_struct *prev,struct mm_struct *next) { -switch_mm (prev, next, current); +switch_mm_irqs_off (prev, next, current); }/* We don't currently use enter_lazy_tlb() for anything */
Signature · 1 line
2.27.0
Re: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Quoted 22 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0
arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
This depends on upstream commit:
d53c3dfb23c4 ("mm: fix exec activate_mm vs TLB shootdown and lazy tlb switching race")
Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
cheers
Re: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Quoted 30 lines · Michael Ellerman
Quoted 22 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MMThis depends on upstream commit:
Quoted 7 lines
d53c3dfb23c4("mm: fix execactivate_mmvs TLB shootdown and lazy tlb switching race")Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
I just sent you a backport of d53c3dfb23c4 for 4.19.
cheers
Greg Kroah-Hartman
maintainerRe: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Quoted 34 lines · Michael Ellerman
Quoted 33 lines · Michael Ellerman
Quoted 22 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MMThis depends on upstream commit:
d53c3dfb23c4("mm: fix execactivate_mmvs TLB shootdown and lazy tlb switching race")
Quoted 6 lines
Quoted 4 lines
Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
I just sent you a backport of
d53c3dfb23c4for 4.19.
I've taken that at the proper part of this series now, thanks!
greg k-h
Same problem for 4.14.y and a concern: Re: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
In reply to Michael EllermanQuoted 35 lines · Michael Ellerman
Quoted 11 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Link:
https://lore.kernel.org/r/[email protected]
Quoted 11 lines
Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MMThis depends on upstream commit:
d53c3dfb23c4("mm: fix execactivate_mmvs TLB shootdown and lazy tlb
switching race")
Quoted 7 lines
Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
cheers
Hi-
This glitch has made its way into 4.14.y ...
[4.14.y] c2bca8712a19 powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
But 4.14.y does not carry the prereq that introduces that config.
That said, I have a more general concern about the new config (in mainline and the stable backports):
[mainline] d53c3dfb23c4 mm: fix exec activate_mm vs TLB shootdown and
lazy tlb switching race
It would seem that the intent is that it should be *only* enabled (currently at least) for arches that will explicitly select it, but the config advice does not make that very clear. Could that new config get an explicit "default n" line?
-Kamal
Sasha Levin
maintainerRe: Same problem for 4.14.y and a concern: Re: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Quoted 42 lines · Kamal Mostafa
Quoted 40 lines · Michael Ellerman
Quoted 11 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Link:
https://lore.kernel.org/r/[email protected]
Quoted 11 lines
Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MMThis depends on upstream commit:
d53c3dfb23c4("mm: fix execactivate_mmvs TLB shootdown and lazy tlb
switching race")Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
Quoted 8 lines
Quoted 2 lines
cheers
Hi-
This glitch has made its way into 4.14.y ...
[4.14.y]c2bca8712a19powerpc: selectARCH_WANT_IRQS_OFF_ACTIVATE_MM
But 4.14.y does not carry the prereq that introduces that config.
I'll queue up the 4.19 backport for 4.14 too, thanks!
Signature · 2 lines
Thanks, Sasha
Re: Same problem for 4.14.y and a concern: Re: [PATCH 4.19 056/191] powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
Quoted 46 lines · Sasha Levin
Quoted 45 lines · Kamal Mostafa
Quoted 42 lines · Michael Ellerman
Quoted 11 lines · Greg Kroah-Hartman
From: Nicholas Piggin <[email protected]>
[ Upstream commit
66acd46080bd9e5ad2be4b0eb1d498d5145d058e]powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing
ARCH_WANT_IRQS_OFF_ACTIVATE_MM.Signed-off-by Nicholas Piggin <[email protected]>Signed-off-by Michael Ellerman <[email protected]>Link:
https://lore.kernel.org/r/[email protected]
Quoted 11 lines
Signed-off-by Sasha Levin <[email protected]>--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)arch/powerpc/Kconfig +0
-0arch/powerpc/Kconfig config PPC · line 155
Quoted 4 lines
arch/powerpc/Kconfig +1
-0@@ -152,6 +152,7 @@ config PPC select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MMThis depends on upstream commit:
d53c3dfb23c4("mm: fix execactivate_mmvs TLB shootdown and lazy tlb
switching race")Which I don't see in 4.19 stable, or in the email thread here.
So this shouldn't be backported to 4.19 unless that commit is also backported.
cheers
Quoted 8 lines
Quoted 5 lines
Hi-
This glitch has made its way into 4.14.y ...
[4.14.y]c2bca8712a19powerpc: selectARCH_WANT_IRQS_OFF_ACTIVATE_MM
But 4.14.y does not carry the prereq that introduces that config.I'll queue up the 4.19 backport for 4.14 too, thanks!
Thanks Sasha.
And nevermind my other concern ...
Quoted 4 lines
Quoted 4 lines
It would seem that the intent is that it should be *only* enabled (currently at least) for arches that will explicitly select it, but the config advice does not make that very clear. Could that new config get an explicit "default n" line?
... I see now that a 'default' isn't necessary; the config only appears for arches which explicitly select it, as intended.
-Kamal
Greg Kroah-Hartman
maintainerPATCH 57/191 sparc64: remove mm_cpumask clearing to fix kthread_use_mm race
In reply to Greg Kroah-HartmanFrom: Nicholas Piggin <[email protected]>
[ Upstream commit bafb056ce27940c9994ea905336aa8f27b4f7275 ]
The de facto (and apparently uncommented) standard for using an mm had, thanks to this code in sparc if nothing else, been that you must have a reference on mm_users *and that reference must have been obtained with mmget()*, i.e., from a thread with a reference to mm_users that had used the mm.
The introduction of mmget_not_zero() in commit d2005e3f41d4 ("userfaultfd: don't pin the user memory in userfaultfd_file_create()") allowed mm_count holders to aoperate on user mappings asynchronously from the actual threads using the mm, but they were not to load those mappings into their TLB (i.e., walking vmas and page tables is okay, kthread_use_mm() is not).
io_uring 2b188cc1bb857 ("Add io_uring IO interface") added code which does a kthread_use_mm() from a mmget_not_zero() refcount.
The problem with this is code which previously assumed mm == current->mm and mm->mm_users == 1 implies the mm will remain single-threaded at least until this thread creates another mm_users reference, has now broken.
arch/sparc/kernel/smp_64.c:
if (atomic_read(&mm->mm_users) == 1) {
cpumask_copy(mm_cpumask(mm), cpumask_of(cpu));
goto local_flush_and_out;
}vs fs/io_uring.c
if (unlikely(!(ctx->flags & IORING_SETUP_SQPOLL) ||
!mmget_not_zero(ctx->sqo_mm)))
return -EFAULT;
kthread_use_mm(ctx->sqo_mm);mmget_not_zero() could come in right after the mm_users == 1 test, then kthread_use_mm() which sets its CPU in the mm_cpumask. That update could be lost if cpumask_copy() occurs afterward.
I propose we fix this by allowing mmget_not_zero() to be a first-class reference, and not have this obscure undocumented and unchecked restriction.
The basic fix for sparc64 is to remove its mm_cpumask clearing code. The optimisation could be effectively restored by sending IPIs to mm_cpumask members and having them remove themselves from mm_cpumask. This is more tricky so I leave it as an exercise for someone with a sparc64 SMP. powerpc has a (currently similarly broken) example.
--- arch/sparc/kernel/smp_64.c | 65 ++++++++------------------------------ 1 file changed, 14 insertions(+), 51 deletions(-)
arch/sparc/kernel/smp_64.c +14 -51
@@ -1039,38 +1039,9 @@ void smp_fetch_global_pmu(void)* are flush_tlb_*() routines, and these run after flush_cache_*() * which performs the flushw. * -* The SMP TLB coherency scheme we use works as follows: -* -* 1) mm->cpu_vm_mask is a bit mask of which cpus an address -* space has (potentially) executed on, this is the heuristic -* we use to avoid doing cross calls. -* -* Also, for flushing from kswapd and also for clones, we -* use cpu_vm_mask as the list of cpus to make run the TLB. -* -* 2) TLB context numbers are shared globally across all processors -* in the system, this allows us to play several games to avoid -* cross calls. -* -* One invariant is that when a cpu switches to a process, and -* that processes tsk->active_mm->cpu_vm_mask does not have the -* current cpu's bit set, that tlb context is flushed locally. -* -* If the address space is non-shared (ie. mm->count == 1) we avoid -* cross calls when we want to flush the currently running process's -* tlb state. This is done by clearing all cpu bits except the current -* processor's in current->mm->cpu_vm_mask and performing the -* flush locally only. This will force any subsequent cpus which run -* this task to flush the context from the local tlb if the process -* migrates to another cpu (again). -* -* 3) For shared address spaces (threads) and swapping we bite the -* bullet for most cases and perform the cross call (but only to -* the cpus listed in cpu_vm_mask). -* -* The performance gain from "optimizing" away the cross call for threads is -* questionable (in theory the big win for threads is the massive sharing of -* address space state across processors). +* mm->cpu_vm_mask is a bit mask of which cpus an address +* space has (potentially) executed on, this is the heuristic +* we use to limit cross calls. */ /* This currently is only used by the hugetlb arch pre-fault
@@ -1080,18 +1051,13 @@ void smp_fetch_global_pmu(void)void smp_flush_tlb_mm (struct mm_struct *mm) {u32 ctx =CTX_HWBITS (mm->context); -int cpu =get_cpu (); -if (atomic_read (&mm->mm_users) ==1 ) { -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); - goto local_flush_and_out; - } +get_cpu ();smp_cross_call_masked (&xcall_flush_tlb_mm, ctx,0 ,0 ,mm_cpumask (mm)); -local_flush_and_out:__flush_tlb_mm (ctx,SECONDARY_CONTEXT );put_cpu ();
@@ -1114,17 +1080,15 @@ void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned long {u32 ctx =CTX_HWBITS (mm->context);struct tlb_pending_info info; -int cpu =get_cpu (); + +get_cpu (); info.ctx = ctx; info.nr = nr; info.vaddrs = vaddrs; -if (mm == current->mm &&atomic_read (&mm->mm_users) ==1 ) -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); -else -smp_call_function_many (mm_cpumask (mm), tlb_pending_func, - &info,1 ); +smp_call_function_many (mm_cpumask (mm), tlb_pending_func, + &info,1 );__flush_tlb_pending (ctx, nr, vaddrs);
@@ -1134,14 +1098,13 @@ void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned longvoid smp_flush_tlb_page (struct mm_struct *mm,unsigned long vaddr) {unsigned long context =CTX_HWBITS (mm->context); -int cpu =get_cpu (); -if (mm == current->mm &&atomic_read (&mm->mm_users) ==1 ) -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); -else -smp_cross_call_masked (&xcall_flush_tlb_page, - context, vaddr,0 , -mm_cpumask (mm)); +get_cpu (); + +smp_cross_call_masked (&xcall_flush_tlb_page, + context, vaddr,0 , +mm_cpumask (mm)); +__flush_tlb_page (context, vaddr);put_cpu ();
Signature · 1 line
2.27.0
Re: [PATCH 4.19 057/191] sparc64: remove mm_cpumask clearing to fix kthread_use_mm race
Hi!
Quoted 3 lines
From: Nicholas Piggin <[email protected]>
[ Upstream commit
bafb056ce27940c9994ea905336aa8f27b4f7275]
...
Quoted 2 lines
io_uring2b188cc1bb857("Addio_uringIO interface") added code which does akthread_use_mm()from ammget_not_zero()refcount.
...
Quoted 5 lines
The basic fix for sparc64 is to remove its
mm_cpumaskclearing code. The optimisation could be effectively restored by sending IPIs tomm_cpumaskmembers and having them remove themselves frommm_cpumask. This is more tricky so I leave it as an exercise for someone with a sparc64 SMP. powerpc has a (currently similarly broken) example.
So this removes optimalization from Sparc, because it clashes with 2b188cc1bb857 ("Add io_uring IO interface"). But that commit is not present in 4.19... so this probably is not good idea.
Best regards,
Pavel
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Greg Kroah-Hartman
maintainerPATCH 58/191 f2fs: add trace exit in exception path
In reply to Greg Kroah-HartmanFrom: Zhang Qilong <[email protected]>
[ Upstream commit 9b66482282888d02832b7d90239e1cdb18e4b431 ]
Missing the trace exit in f2fs_sync_dirty_inodes
--- fs/f2fs/checkpoint.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/f2fs/checkpoint.c +5 -1
@@ -1003,8 +1003,12 @@ int f2fs_sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type)get_pages (sbi, is_dir ? F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA)); retry: -if (unlikely (f2fs_cp_error (sbi))) +if (unlikely (f2fs_cp_error (sbi))) { +trace_f2fs_sync_dirty_inodes_exit (sbi->sb, is_dir, +get_pages (sbi, is_dir ? +F2FS_DIRTY_DENTS :F2FS_DIRTY_DATA ));return -EIO ; + }spin_lock (&sbi->inode_lock[type]);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 59/191 f2fs: fix uninit-value in f2fs_lookup
In reply to Greg Kroah-HartmanFrom: Chao Yu <[email protected]>
[ Upstream commit 6d7ab88a98c1b7a47c228f8ffb4f44d631eaf284 ]
As syzbot reported:
Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x21c/0x280 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:122 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:219 f2fs_lookup+0xe05/0x1a80 fs/f2fs/namei.c:503 lookup_open fs/namei.c:3082 [inline] open_last_lookups fs/namei.c:3177 [inline] path_openat+0x2729/0x6a90 fs/namei.c:3365 do_filp_open+0x2b8/0x710 fs/namei.c:3395 do_sys_openat2+0xa88/0x1140 fs/open.c:1168 do_sys_open fs/open.c:1184 [inline] __do_compat_sys_openat fs/open.c:1242 [inline] __se_compat_sys_openat+0x2a4/0x310 fs/open.c:1240 __ia32_compat_sys_openat+0x56/0x70 fs/open.c:1240 do_syscall_32_irqs_on arch/x86/entry/common.c:80 [inline] __do_fast_syscall_32+0x129/0x180 arch/x86/entry/common.c:139 do_fast_syscall_32+0x6a/0xc0 arch/x86/entry/common.c:162 do_SYSENTER_32+0x73/0x90 arch/x86/entry/common.c:205 entry_SYSENTER_compat_after_hwframe+0x4d/0x5c
In f2fs_lookup(), @res_page could be used before being initialized, because in __f2fs_find_entry(), once F2FS_I(dir)->i_current_depth was been fuzzed to zero, then @res_page will never be initialized, causing this kmsan warning, relocating @res_page initialization place to fix this bug.
--- fs/f2fs/dir.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
fs/f2fs/dir.c +3 -5
@@ -208,16 +208,15 @@ struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir,unsigned int max_depth;unsigned int level; + *res_page =NULL ; +if (f2fs_has_inline_dentry (dir)) { - *res_page =NULL ; de =f2fs_find_in_inline_dir (dir, fname, res_page); goto out; } -if (npages ==0 ) { - *res_page =NULL ; +if (npages ==0 ) goto out; - } max_depth =F2FS_I (dir)->i_current_depth;if (unlikely (max_depth >MAX_DIR_HASH_DEPTH )) {
@@ -229,7 +228,6 @@ struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir, }for (level =0 ; level < max_depth; level++) { - *res_page =NULL ; de =find_in_level (dir, level, fname, res_page);if (de ||IS_ERR (*res_page))break ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 60/191 f2fs: fix to check segment boundary during SIT page readahead
In reply to Greg Kroah-HartmanFrom: Chao Yu <[email protected]>
[ Upstream commit 6a257471fa42c8c9c04a875cd3a2a22db148e0f0 ]
As syzbot reported:
kernel BUG at fs/f2fs/segment.h:657!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 16220 Comm: syz-executor.0 Not tainted 5.9.0-rc5-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:f2fs_ra_meta_pages+0xa51/0xdc0 fs/f2fs/segment.h:657 Call Trace: build_sit_entries fs/f2fs/segment.c:4195 [inline] f2fs_build_segment_manager+0x4b8a/0xa3c0 fs/f2fs/segment.c:4779 f2fs_fill_super+0x377d/0x6b80 fs/f2fs/super.c:3633 mount_bdev+0x32e/0x3f0 fs/super.c:1417 legacy_get_tree+0x105/0x220 fs/fs_context.c:592 vfs_get_tree+0x89/0x2f0 fs/super.c:1547 do_new_mount fs/namespace.c:2875 [inline] path_mount+0x1387/0x2070 fs/namespace.c:3192 do_mount fs/namespace.c:3205 [inline] __do_sys_mount fs/namespace.c:3413 [inline] __se_sys_mount fs/namespace.c:3390 [inline] __x64_sys_mount+0x27f/0x300 fs/namespace.c:3390 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9
@blkno in f2fs_ra_meta_pages could exceed max segment count, causing panic in following sanity check in current_sit_addr(), add check condition to avoid this issue.
--- fs/f2fs/checkpoint.c | 2 ++ 1 file changed, 2 insertions(+)
fs/f2fs/checkpoint.c +2 -0
@@ -218,6 +218,8 @@ int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, blkno * NAT_ENTRY_PER_BLOCK);break ;case META_SIT : +if (unlikely (blkno >=TOTAL_SEGS (sbi))) + goto out;/* get sit block addr */ fio.new_blkaddr =current_sit_addr (sbi, blkno *SIT_ENTRY_PER_BLOCK );
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 61/191 um: change sigio_spinlock to a mutex
In reply to Greg Kroah-HartmanFrom: Johannes Berg <[email protected]>
[ Upstream commit f2d05059e15af3f70502074f4e3a504530af504a ]
Lockdep complains at boot:
=============================
[ BUG: Invalid wait context ]
5.7.0-05093-g46d91ecd597b #98 Not tainted
-----------------------------
swapper/1 is trying to lock:0000000060931b98 (&desc[i].request_mutex){+.+.}-{3:3}, at: __setup_irq+0x11d/0x623 other info that might help us debug this:
context-{4:4}
1 lock held by swapper/1:
#0: 000000006074fed8 (sigio_spinlock){+.+.}-{2:2}, at: sigio_lock+0x1a/0x1c
stack backtrace: CPU: 0 PID: 1 Comm: swapper Not tainted 5.7.0-05093-g46d91ecd597b #98 Stack: 7fa4fab0 6028dfd1 0000002a 6008bea5 7fa50700 7fa50040 7fa4fac0 6028e016 7fa4fb50 6007f6da 60959c18 00000000 Call Trace: [<60023a0e>] show_stack+0x13b/0x155 [<6028e016>] dump_stack+0x2a/0x2c [<6007f6da>] __lock_acquire+0x515/0x15f2 [<6007eb50>] lock_acquire+0x245/0x273 [<6050d9f1>] __mutex_lock+0xbd/0x325 [<6050dc76>] mutex_lock_nested+0x1d/0x1f [<6008e27e>] __setup_irq+0x11d/0x623 [<6008e8ed>] request_threaded_irq+0x169/0x1a6 [<60021eb0>] um_request_irq+0x1ee/0x24b [<600234ee>] write_sigio_irq+0x3b/0x76 [<600383ca>] sigio_broken+0x146/0x2e4 [<60020bd8>] do_one_initcall+0xde/0x281
Because we hold sigio_spinlock and then get into requesting an interrupt with a mutex.
Change the spinlock to a mutex to avoid that.
--- arch/um/kernel/sigio.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
arch/um/kernel/sigio.c +3 -3
@@ -36,14 +36,14 @@ int write_sigio_irq(int fd) }/* These are called from os-Linux/sigio.c to protect its pollfds arrays. */ -static DEFINE_SPINLOCK (sigio_spinlock); +static DEFINE_MUTEX (sigio_mutex);void sigio_lock (void ) { -spin_lock (&sigio_spinlock); +mutex_lock (&sigio_mutex); }void sigio_unlock (void ) { -spin_unlock (&sigio_spinlock); +mutex_unlock (&sigio_mutex); }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 62/191 ARM: 8997/2: hw_breakpoint: Handle inexact watchpoint addresses
In reply to Greg Kroah-HartmanFrom: Douglas Anderson <[email protected]>
[ Upstream commit 22c9e58299e5f18274788ce54c03d4fb761e3c5d ]
This is commit fdfeff0f9e3d ("arm64: hw_breakpoint: Handle inexact watchpoint addresses") but ported to arm32, which has the same problem.
This problem was found by Android CTS tests, notably the "watchpoint_imprecise" test [1]. I tested locally against a copycat (simplified) version of the test though.
[1] https://android.googlesource.com/platform/bionic/+/master/tests/sys_ptrace_test.cpp
--- arch/arm/kernel/hw_breakpoint.c | 100 +++++++++++++++++++++++--------- 1 file changed, 72 insertions(+), 28 deletions(-)
arch/arm/kernel/hw_breakpoint.c +72 -28
@@ -688,6 +688,40 @@ static void disable_single_step(struct perf_event *bp)arch_install_hw_breakpoint (bp); } +/* +* Arm32 hardware does not always report a watchpoint hit address that matches +* one of the watchpoints set. It can also report an address "near" the +* watchpoint if a single instruction access both watched and unwatched +* addresses. There is no straight-forward way, short of disassembling the +* offending instruction, to map that address back to the watchpoint. This +* function computes the distance of the memory access from the watchpoint as a +* heuristic for the likelyhood that a given access triggered the watchpoint. +* +* See this same function in the arm64 platform code, which has the same +* problem. +* +* The function returns the distance of the address from the bytes watched by +* the watchpoint. In case of an exact match, it returns 0. +*/ +static u32 get_distance_from_watchpoint (unsigned long addr,u32 val, +struct arch_hw_breakpoint_ctrl *ctrl) +{ +u32 wp_low, wp_high; +u32 lens, lene; + + lens =__ffs (ctrl->len); + lene =__fls (ctrl->len); + + wp_low = val + lens; + wp_high = val + lene; +if (addr < wp_low) +return wp_low - addr; +else if (addr > wp_high) +return addr - wp_high; +else +return 0 ; +} +static int watchpoint_fault_on_uaccess (struct pt_regs *regs,struct arch_hw_breakpoint *info) {
@@ -697,23 +731,25 @@ static int watchpoint_fault_on_uaccess(struct pt_regs *regs,static void watchpoint_handler (unsigned long addr,unsigned int fsr,struct pt_regs *regs) { -int i, access; -u32 val, ctrl_reg, alignment_mask; +int i, access, closest_match =0 ; +u32 min_dist =-1 , dist; +u32 val, ctrl_reg;struct perf_event *wp, **slots;struct arch_hw_breakpoint *info;struct arch_hw_breakpoint_ctrl ctrl; slots =this_cpu_ptr (wp_on_reg); +/* +* Find all watchpoints that match the reported address. If no exact +* match is found. Attribute the hit to the closest watchpoint. +*/ +rcu_read_lock ();for (i =0 ; i < core_num_wrps; ++i) { -rcu_read_lock (); - wp = slots[i]; -if (wp ==NULL ) - goto unlock; +continue ; - info =counter_arch_bp (wp);/* * The DFAR is an unknown value on debug architectures prior * to 7.1. Since we only allow a single watchpoint on these
@@ -722,33 +758,31 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,*/ if (debug_arch <ARM_DEBUG_ARCH_V7_1 ) {BUG_ON (i >0 ); + info =counter_arch_bp (wp); info->trigger = wp->attr.bp_addr; }else { -if (info->ctrl.len ==ARM_BREAKPOINT_LEN_8 ) - alignment_mask =0x7 ; -else - alignment_mask =0x3 ; - -/* Check if the watchpoint value matches. */ - val =read_wb_reg (ARM_BASE_WVR + i); -if (val != (addr & ~alignment_mask)) - goto unlock; - -/* Possible match, check the byte address select. */ - ctrl_reg =read_wb_reg (ARM_BASE_WCR + i); -decode_ctrl_reg (ctrl_reg, &ctrl); -if (!((1 << (addr & alignment_mask)) & ctrl.len)) - goto unlock; -/* Check that the access type matches. */ if (debug_exception_updates_fsr ()) { access = (fsr &ARM_FSR_ACCESS_MASK ) ?HW_BREAKPOINT_W :HW_BREAKPOINT_R ;if (!(access &hw_breakpoint_type (wp))) - goto unlock; +continue ; } + val =read_wb_reg (ARM_BASE_WVR + i); + ctrl_reg =read_wb_reg (ARM_BASE_WCR + i); +decode_ctrl_reg (ctrl_reg, &ctrl); + dist =get_distance_from_watchpoint (addr, val, &ctrl); +if (dist < min_dist) { + min_dist = dist; + closest_match = i; + } +/* Is this an exact match? */ +if (dist !=0 ) +continue ; +/* We have a winner. */ + info =counter_arch_bp (wp); info->trigger = addr; }
@@ -770,13 +804,23 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,* we can single-step over the watchpoint trigger. */ if (!is_default_overflow_handler (wp)) - goto unlock; - +continue ; step:enable_single_step (wp,instruction_pointer (regs)); -unlock: -rcu_read_unlock (); } + +if (min_dist >0 && min_dist !=-1 ) { +/* No exact match found. */ + wp = slots[closest_match]; + info =counter_arch_bp (wp); + info->trigger = addr; +pr_debug ("watchpoint fired: address = 0x%x\n" , info->trigger); +perf_bp_event (wp, regs); +if (is_default_overflow_handler (wp)) +enable_single_step (wp,instruction_pointer (regs)); + } + +rcu_read_unlock (); }static void watchpoint_single_step_handler (unsigned long pc)
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 63/191 power: supply: bq27xxx: report "not charging" on all types
In reply to Greg Kroah-HartmanFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 7bf738ba110722b63e9dc8af760d3fb2aef25593 ]
Commit 6f24ff97e323 ("power: supply: bq27xxx_battery: Add the BQ27Z561 Battery monitor") and commit d74534c27775 ("power:bq27xxx_battery: Add support for additional bq27xxx family devices") added support for new device types by copying most of the code and adding necessary quirks.
However they did not copy the code in bq27xxx_battery_status() responsible for returning POWER_SUPPLY_STATUS_NOT_CHARGING.
Unify the bq27xxx_battery_status() so for all types when charger is supplied, it will return "not charging" status.
--- drivers/power/supply/bq27xxx_battery.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/power/supply/bq27xxx_battery.c +4 -2
@@ -1680,8 +1680,6 @@ static int bq27xxx_battery_status(struct bq27xxx_device_info *di, status =POWER_SUPPLY_STATUS_FULL ; else if (di->cache.flags & BQ27000_FLAG_CHGS) status =POWER_SUPPLY_STATUS_CHARGING ; - else if (power_supply_am_i_supplied(di->bat) > 0) - status =POWER_SUPPLY_STATUS_NOT_CHARGING ; else status =POWER_SUPPLY_STATUS_DISCHARGING ; }else {
@@ -1693,6 +1691,10 @@ static int bq27xxx_battery_status(struct bq27xxx_device_info *di, status =POWER_SUPPLY_STATUS_CHARGING ; } +if ((status ==POWER_SUPPLY_STATUS_DISCHARGING ) && + (power_supply_am_i_supplied (di->bat) >0 )) + status =POWER_SUPPLY_STATUS_NOT_CHARGING ; + val->intval = status;return 0 ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 64/191 xfs: fix realtime bitmap/summary file truncation when growing rt volume
In reply to Greg Kroah-HartmanFrom: Darrick J. Wong <[email protected]>
[ Upstream commit f4c32e87de7d66074d5612567c5eac7325024428 ]
The realtime bitmap and summary files are regular files that are hidden away from the directory tree. Since they're regular files, inode inactivation will try to purge what it thinks are speculative preallocations beyond the incore size of the file. Unfortunately, xfs_growfs_rt forgets to update the incore size when it resizes the inodes, with the result that inactivating the rt inodes at unmount time will cause their contents to be truncated.
Fix this by updating the incore size when we change the ondisk size as part of updating the superblock. Note that we don't do this when we're allocating blocks to the rt inodes because we actually want those blocks to get purged if the growfs fails.
This fixes corruption complaints from the online rtsummary checker when running xfs/233. Since that test requires rmap, one can also trigger this by growing an rt volume, cycling the mount, and creating rt files.
--- fs/xfs/xfs_rtalloc.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
fs/xfs/xfs_rtalloc.c +8 -2
@@ -998,10 +998,13 @@ xfs_growfs_rt(xfs_ilock (mp->m_rbmip,XFS_ILOCK_EXCL );xfs_trans_ijoin (tp, mp->m_rbmip,XFS_ILOCK_EXCL );/* -* Update the bitmap inode's size. +* Update the bitmap inode's size ondisk and incore. We need +* to update the incore size so that inode inactivation won't +* punch what it thinks are "posteof" blocks. */ mp->m_rbmip->i_d.di_size = nsbp->sb_rbmblocks * nsbp->sb_blocksize; +i_size_write (VFS_I (mp->m_rbmip), mp->m_rbmip->i_d.di_size);xfs_trans_log_inode (tp, mp->m_rbmip,XFS_ILOG_CORE );/* * Get the summary inode into the transaction.
@@ -1009,9 +1012,12 @@ xfs_growfs_rt(xfs_ilock (mp->m_rsumip,XFS_ILOCK_EXCL );xfs_trans_ijoin (tp, mp->m_rsumip,XFS_ILOCK_EXCL );/* -* Update the summary inode's size. +* Update the summary inode's size. We need to update the +* incore size so that inode inactivation won't punch what it +* thinks are "posteof" blocks. */ mp->m_rsumip->i_d.di_size = nmp->m_rsumsize; +i_size_write (VFS_I (mp->m_rsumip), mp->m_rsumip->i_d.di_size);xfs_trans_log_inode (tp, mp->m_rsumip,XFS_ILOG_CORE );/* * Copy summary data from old to new sizes.
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 65/191 video: fbdev: pvr2fb: initialize variables
In reply to Greg Kroah-HartmanFrom: Tom Rix <[email protected]>
[ Upstream commit 8e1ba47c60bcd325fdd097cd76054639155e5d2e ]
clang static analysis reports this repesentative error
pvr2fb.c:1049:2: warning: 1st function call argument
is an uninitialized value [core.CallAndMessage]
if (*cable_arg)
^~~~~~~~~~~~~~~Problem is that cable_arg depends on the input loop to set the cable_arg[0]. If it does not, then some random value from the stack is used.
A similar problem exists for output_arg.
So initialize cable_arg and output_arg.
--- drivers/video/fbdev/pvr2fb.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/video/fbdev/pvr2fb.c +2 -0
@@ -1027,6 +1027,8 @@ static int __init pvr2fb_setup(char *options)if (!options || !*options)return 0 ; + cable_arg[0 ] = output_arg[0 ] =0 ; +while ((this_opt =strsep (&options,"," ))) {if (!*this_opt)continue ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 66/191 ath10k: start recovery process when payload length exceeds max htc length for sdio
In reply to Greg Kroah-HartmanFrom: Wen Gong <[email protected]>
[ Upstream commit 2fd3c8f34d08af0a6236085f9961866ad92ef9ec ]
When simulate random transfer fail for sdio write and read, it happened "payload length exceeds max htc length" and recovery later sometimes.
Test steps:
1. Add config and update kernel:CONFIG_FAIL_MMC_REQUEST=yCONFIG_FAULT_INJECTION=yCONFIG_FAULT_INJECTION_DEBUG_FS=y
2. Run simulate fail:
cd /sys/kernel/debug/mmc1/fail_mmc_request
echo 10 > probability
echo 10 > times # repeat until hitting issues
3. It happened payload length exceeds max htc length.
[ 199.935506] ath10k_sdio mmc1:0001:1: payload length 57005 exceeds max htc length: 4088 .... [ 264.990191] ath10k_sdio mmc1:0001:1: payload length 57005 exceeds max htc length: 4088
4. after some time, such as 60 seconds, it start recovery which triggered by wmi command timeout for periodic scan.
[ 269.229232] ieee80211 phy0: Hardware restart was requested [ 269.734693] ath10k_sdio mmc1:0001:1: device successfully recovered
The simulate fail of sdio is not a real sdio transter fail, it only set an error status in mmc_should_fail_request after the transfer end, actually the transfer is success, then sdio_io_rw_ext_helper will return error status and stop transfer the left data. For example, the really RX len is 286 bytes, then it will split to 2 blocks in sdio_io_rw_ext_helper, one is 256 bytes, left is 30 bytes, if the first 256 bytes get an error status by mmc_should_fail_request,then the left 30 bytes will not read in this RX operation. Then when the next RX arrive, the left 30 bytes will be considered as the header of the read, the top 4 bytes of the 30 bytes will be considered as lookaheads, but actually the 4 bytes is not the lookaheads, so the len from this lookaheads is not correct, it exceeds max htc length 4088 sometimes. When happened exceeds, the buffer chain is not matched between firmware and ath10k, then it need to start recovery ASAP. Recently then recovery will be started by wmi command timeout, but it will be long time later, for example, it is 60+ seconds later from the periodic scan, if it does not have periodic scan, it will be longer.
Start recovery when it happened "payload length exceeds max htc length" will be reasonable.
This patch only effect sdio chips.
Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.
--- drivers/net/wireless/ath/ath10k/sdio.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/net/wireless/ath/ath10k/sdio.c +4 -0
@@ -561,6 +561,10 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,le16_to_cpu (htc_hdr->len), ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH); ret = -ENOMEM ; + +queue_work (ar->workqueue, &ar->restart_work); +ath10k_warn (ar,"exceeds length, start recovery\n" ); + goto err; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 67/191 ath10k: fix VHT NSS calculation when STBC is enabled
In reply to Greg Kroah-HartmanFrom: Sathishkumar Muruganandam <[email protected]>
[ Upstream commit 99f41b8e43b8b4b31262adb8ac3e69088fff1289 ]
When STBC is enabled, NSTS_SU value need to be accounted for VHT NSS calculation for SU case.
Without this fix, 1SS + STBC enabled case was reported wrongly as 2SS in radiotap header on monitor mode capture.
Tested-on: QCA9984 10.4-3.10-00047
--- drivers/net/wireless/ath/ath10k/htt_rx.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath10k/htt_rx.c +7 -1
@@ -767,6 +767,7 @@ static void ath10k_htt_rx_h_rates(struct ath10k *ar,u8 preamble =0 ;u8 group_id;u32 info1, info2, info3; +u32 stbc, nsts_su; info1 =__le32_to_cpu (rxd->ppdu_start.info1); info2 =__le32_to_cpu (rxd->ppdu_start.info2);
@@ -811,11 +812,16 @@ static void ath10k_htt_rx_h_rates(struct ath10k *ar,*/ bw = info2 &3 ; sgi = info3 &1 ; + stbc = (info2 >>3 ) &1 ; group_id = (info2 >>4 ) &0x3F ;if (GROUP_ID_IS_SU_MIMO (group_id)) { mcs = (info3 >>4 ) &0x0F ; - nss = ((info2 >>10 ) &0x07 ) +1 ; + nsts_su = ((info2 >>10 ) &0x07 ); +if (stbc) + nss = (nsts_su >>2 ) +1 ; +else + nss = (nsts_su +1 ); }else {/* Hardware doesn't decode VHT-SIG-B into Rx descriptor * so it's impossible to decode MCS. Also since
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 68/191 drm/brige/megachips: Add checking if ge_b850v3_lvds_init() is working correctly
In reply to Greg Kroah-HartmanFrom: Nadezda Lutovinova <[email protected]>
[ Upstream commit f688a345f0d7a6df4dd2aeca8e4f3c05e123a0ee ]
If ge_b850v3_lvds_init() does not allocate memory for ge_b850v3_lvds_ptr, then a null pointer dereference is accessed.
The patch adds checking of the return value of ge_b850v3_lvds_init().
Found by Linux Driver Verification project (linuxtesting.org).
--- .../gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c +10 -2
@@ -306,8 +306,12 @@ static int stdp4028_ge_b850v3_fw_probe(struct i2c_client *stdp4028_i2c, const struct i2c_device_id *id) {struct device *dev = &stdp4028_i2c->dev; +int ret; + + ret =ge_b850v3_lvds_init (dev); -ge_b850v3_lvds_init (dev); +if (ret) +return ret; ge_b850v3_lvds_ptr->stdp4028_i2c = stdp4028_i2c;i2c_set_clientdata (stdp4028_i2c, ge_b850v3_lvds_ptr);
@@ -365,8 +369,12 @@ static int stdp2690_ge_b850v3_fw_probe(struct i2c_client *stdp2690_i2c, const struct i2c_device_id *id) {struct device *dev = &stdp2690_i2c->dev; +int ret; + + ret =ge_b850v3_lvds_init (dev); -ge_b850v3_lvds_init (dev); +if (ret) +return ret; ge_b850v3_lvds_ptr->stdp2690_i2c = stdp2690_i2c;i2c_set_clientdata (stdp2690_i2c, ge_b850v3_lvds_ptr);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 69/191 media: videodev2.h: RGB BT2020 and HSV are always full range
In reply to Greg Kroah-HartmanFrom: Hans Verkuil <[email protected]>
[ Upstream commit b305dfe2e93434b12d438434461b709641f62af4 ]
The default RGB quantization range for BT.2020 is full range (just as for all the other RGB pixel encodings), not limited range.
Update the V4L2_MAP_QUANTIZATION_DEFAULT macro and documentation accordingly.
Also mention that HSV is always full range and cannot be limited range.
When RGB BT2020 was introduced in V4L2 it was not clear whether it should be limited or full range, but full range is the right (and consistent) choice.
--- .../media/uapi/v4l/colorspaces-defs.rst | 9 ++++----- .../media/uapi/v4l/colorspaces-details.rst | 5 ++--- include/uapi/linux/videodev2.h | 17 ++++++++--------- 3 files changed, 14 insertions(+), 17 deletions(-)
Documentation/media/uapi/v4l/colorspaces-defs.rst +4 -5
@@ -29,8 +29,7 @@ whole range, 0-255, dividing the angular value by 1.41. The enum :c:type:`v4l2_hsv_encoding` specifies which encoding is used. .. note:: The default R'G'B' quantization is full range for all - colorspaces except for BT.2020 which uses limited range R'G'B' - quantization. + colorspaces. HSV formats are always full range. .. tabularcolumns:: |p{6.0cm}|p{11.5cm}|
@@ -162,8 +161,8 @@ whole range, 0-255, dividing the angular value by 1.41. The enum - Details * - ``V4L2_QUANTIZATION_DEFAULT`` - Use the default quantization encoding as defined by the - colorspace. This is always full range for R'G'B' (except for the - BT.2020 colorspace) and HSV. It is usually limited range for Y'CbCr. + colorspace. This is always full range for R'G'B' and HSV. + It is usually limited range for Y'CbCr. * - ``V4L2_QUANTIZATION_FULL_RANGE`` - Use the full range quantization encoding. I.e. the range [0…1] is mapped to [0…255] (with possible clipping to [1…254] to avoid the
@@ -173,4 +172,4 @@ whole range, 0-255, dividing the angular value by 1.41. The enum * - ``V4L2_QUANTIZATION_LIM_RANGE`` - Use the limited range quantization encoding. I.e. the range [0…1] is mapped to [16…235]. Cb and Cr are mapped from [-0.5…0.5] to - [16…240]. + [16…240]. Limited Range cannot be used with HSV.
Documentation/media/uapi/v4l/colorspaces-details.rst +2 -3
@@ -370,9 +370,8 @@ Colorspace BT.2020 (V4L2_COLORSPACE_BT2020) The :ref:`itu2020` standard defines the colorspace used by Ultra-high definition television (UHDTV). The default transfer function is ``V4L2_XFER_FUNC_709``. The default Y'CbCr encoding is -``V4L2_YCBCR_ENC_BT2020``. The default R'G'B' quantization is limited -range (!), and so is the default Y'CbCr quantization. The chromaticities -of the primary colors and the white reference are: +``V4L2_YCBCR_ENC_BT2020``. The default Y'CbCr quantization is limited range. +The chromaticities of the primary colors and the white reference are:
include/uapi/linux/videodev2.h +8 -9
@@ -362,9 +362,9 @@ enum v4l2_hsv_encoding {enum v4l2_quantization {/* -* The default for R'G'B' quantization is always full range, except -* for the BT2020 colorspace. For Y'CbCr the quantization is always -* limited range, except for COLORSPACE_JPEG: this is full range. +* The default for R'G'B' quantization is always full range. +* For Y'CbCr the quantization is always limited range, except +* for COLORSPACE_JPEG: this is full range. */ V4L2_QUANTIZATION_DEFAULT =0 ,V4L2_QUANTIZATION_FULL_RANGE =1 ,
@@ -373,14 +373,13 @@ enum v4l2_quantization {/* * Determine how QUANTIZATION_DEFAULT should map to a proper quantization. -* This depends on whether the image is RGB or not, the colorspace and the -* Y'CbCr encoding. +* This depends on whether the image is RGB or not, the colorspace. +* The Y'CbCr encoding is not used anymore, but is still there for backwards +* compatibility. */ #define V4L2_MAP_QUANTIZATION_DEFAULT (is_rgb_or_hsv, colsp, ycbcr_enc) \ - (((is_rgb_or_hsv) && (colsp) == V4L2_COLORSPACE_BT2020) ? \ - V4L2_QUANTIZATION_LIM_RANGE : \ - (((is_rgb_or_hsv) || (colsp) == V4L2_COLORSPACE_JPEG) ? \ - V4L2_QUANTIZATION_FULL_RANGE : V4L2_QUANTIZATION_LIM_RANGE)) + (((is_rgb_or_hsv) || (colsp) == V4L2_COLORSPACE_JPEG) ? \ + V4L2_QUANTIZATION_FULL_RANGE : V4L2_QUANTIZATION_LIM_RANGE)/* * Deprecated names for opRGB colorspace (IEC 61966-2-5)
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 70/191 media: platform: Improve queue set up flow for bug fixing
In reply to Greg Kroah-HartmanFrom: Xia Jiang <[email protected]>
[ Upstream commit 5095a6413a0cf896ab468009b6142cb0fe617e66 ]
Add checking created buffer size follow in mtk_jpeg_queue_setup().
--- drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c +7 -0
@@ -579,6 +579,13 @@ static int mtk_jpeg_queue_setup(struct vb2_queue *q,if (!q_data)return -EINVAL ; +if (*num_planes) { +for (i =0 ; i < *num_planes; i++) +if (sizes[i] < q_data->sizeimage[i]) +return -EINVAL ; +return 0 ; + } + *num_planes = q_data->fmt->colplanes;for (i =0 ; i < q_data->fmt->colplanes; i++) { sizes[i] = q_data->sizeimage[i];
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 71/191 usb: typec: tcpm: During PR_SWAP, source caps should be sent only after tSwapSourceStart
In reply to Greg Kroah-HartmanFrom: Badhri Jagan Sridharan <[email protected]>
[ Upstream commit 6bbe2a90a0bb4af8dd99c3565e907fe9b5e7fd88 ]
The patch addresses the compliance test failures while running TD.PD.CP.E3, TD.PD.CP.E4, TD.PD.CP.E5 of the "Deterministic PD Compliance MOI" test plan published in https://www.usb.org/usbc.
For a product to be Type-C compliant, it's expected that these tests are run on usb.org certified Type-C compliance tester as mentioned in
https://www.usb.org/usbc.
The purpose of the tests TD.PD.CP.E3, TD.PD.CP.E4, TD.PD.CP.E5 is to verify the PR_SWAP response of the device. While doing so, the test asserts that Source Capabilities message is NOT received from the test device within tSwapSourceStart min (20 ms) from the time the last bit of GoodCRC corresponding to the RS_RDY message sent by the UUT was sent. If it does then the test fails.
This is in line with the requirements from the USB Power Delivery Specification Revision 3.0, Version 1.2:
"6.6.8.1 SwapSourceStartTimer
The SwapSourceStartTimer Shall be used by the new Source, after a Power Role Swap or Fast Role Swap, to ensure that it does not send Source_Capabilities Message before the new Sink is ready to receive theSource_Capabilities Message. The new Source Shall Not send the Source_Capabilities Message earlier than tSwapSourceStart after the last bit of the EOP of GoodCRC Message sent in response to the PS_RDY Message sent by the new Source indicating that its power supply is ready."
The patch makes sure that TCPM does not send the Source_Capabilities Message within tSwapSourceStart(20ms) by transitioning into SRC_STARTUP only after tSwapSourceStart(20ms).
--- drivers/usb/typec/tcpm.c | 2 +- include/linux/usb/pd.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/usb/typec/tcpm.c +1 -1
@@ -3486,7 +3486,7 @@ static void run_state_machine(struct tcpm_port *port)*/ tcpm_set_pwr_role (port,TYPEC_SOURCE );tcpm_pd_send_control (port,PD_CTRL_PS_RDY ); -tcpm_set_state (port,SRC_STARTUP ,0 ); +tcpm_set_state (port,SRC_STARTUP ,PD_T_SWAP_SRC_START );break ;case VCONN_SWAP_ACCEPT :
include/linux/usb/pd.h +1 -0
@@ -451,6 +451,7 @@ static inline unsigned int rdo_max_power(u32 rdo)#define PD_T_ERROR_RECOVERY 100/* minimum 25 is insufficient */ #define PD_T_SRCSWAPSTDBY 625/* Maximum of 650ms */ #define PD_T_NEWSRC 250/* Maximum of 275ms */ +#define PD_T_SWAP_SRC_START 20/* Minimum of 20ms */ #define PD_T_DRP_TRY 100/* 75 - 150 ms */ #define PD_T_DRP_TRYWAIT 600/* 400 - 800 ms */
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 72/191 media: tw5864: check status of tw5864_frameinterval_get
In reply to Greg Kroah-HartmanFrom: Tom Rix <[email protected]>
[ Upstream commit 780d815dcc9b34d93ae69385a8465c38d423ff0f ]
clang static analysis reports this problem
tw5864-video.c:773:32: warning: The left expression of the compound
assignment is an uninitialized value.
The computed value will also be garbage
fintv->stepwise.max.numerator *= std_max_fps;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^stepwise.max is set with frameinterval, which comes from
ret = tw5864_frameinterval_get(input, &frameinterval); fintv->stepwise.step = frameinterval; fintv->stepwise.min = frameinterval; fintv->stepwise.max = frameinterval; fintv->stepwise.max.numerator *= std_max_fps;
When tw5864_frameinterval_get() fails, frameinterval is not set. So check the status and fix another similar problem.
--- drivers/media/pci/tw5864/tw5864-video.c | 6 ++++++ 1 file changed, 6 insertions(+)
drivers/media/pci/tw5864/tw5864-video.c +6 -0
@@ -776,6 +776,9 @@ static int tw5864_enum_frameintervals(struct file *file, void *priv, fintv->type =V4L2_FRMIVAL_TYPE_STEPWISE ; ret =tw5864_frameinterval_get (input, &frameinterval); +if (ret) +return ret; + fintv->stepwise.step = frameinterval; fintv->stepwise.min = frameinterval; fintv->stepwise.max = frameinterval;
@@ -794,6 +797,9 @@ static int tw5864_g_parm(struct file *file, void *priv, cp->capability =V4L2_CAP_TIMEPERFRAME ; ret =tw5864_frameinterval_get (input, &cp->timeperframe); +if (ret) +return ret; + cp->timeperframe.numerator *= input->frame_interval; cp->capturemode =0 ; cp->readbuffers =2 ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 73/191 media: imx274: fix frame interval handling
In reply to Greg Kroah-HartmanFrom: Hans Verkuil <[email protected]>
[ Upstream commit 49b20d981d723fae5a93843c617af2b2c23611ec ]
1) the numerator and/or denominator might be 0, in that case fall back to the default frame interval. This is per the spec and this caused a v4l2-compliance failure.
2) the updated frame interval wasn't returned in the s_frame_interval subdev op.
--- drivers/media/i2c/imx274.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/media/i2c/imx274.c +5 -3
@@ -1239,6 +1239,8 @@ static int imx274_s_frame_interval(struct v4l2_subdev *sd, ret =imx274_set_frame_interval (imx274, fi->interval);if (!ret) { + fi->interval = imx274->frame_interval; +/* * exposure time range is decided by frame interval * need to update it after frame interval changes
@@ -1760,9 +1762,9 @@ static int imx274_set_frame_interval(struct stimx274 *priv, __func__, frame_interval.numerator, frame_interval.denominator); -if (frame_interval.numerator ==0 ) { - err = -EINVAL ; - goto fail; +if (frame_interval.numerator ==0 || frame_interval.denominator ==0 ) { + frame_interval.denominator =IMX274_DEF_FRAME_RATE ; + frame_interval.numerator =1 ; } req_frame_rate = (u32)(frame_interval.denominator
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 74/191 mmc: via-sdmmc: Fix data race bug
In reply to Greg Kroah-HartmanFrom: Madhuparna Bhowmik <[email protected]>
[ Upstream commit 87d7ad089b318b4f319bf57f1daa64eb6d1d10ad ]
via_save_pcictrlreg() should be called with host->lock held as it writes to pm_pcictrl_reg, otherwise there can be a race condition between via_sd_suspend() and via_sdc_card_detect(). The same pattern is used in the function via_reset_pcictrl() as well, where via_save_pcictrlreg() is called with host->lock held.
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/mmc/host/via-sdmmc.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/mmc/host/via-sdmmc.c +3 -0
@@ -1273,11 +1273,14 @@ static void via_init_sdc_pm(struct via_crdr_mmc_host *host)static int via_sd_suspend (struct pci_dev *pcidev,pm_message_t state) {struct via_crdr_mmc_host *host; +unsigned long flags; host =pci_get_drvdata (pcidev); +spin_lock_irqsave (&host->lock, flags);via_save_pcictrlreg (host);via_save_sdcreg (host); +spin_unlock_irqrestore (&host->lock, flags);pci_save_state (pcidev);pci_enable_wake (pcidev,pci_choose_state (pcidev, state),0 );
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 75/191 drm/bridge/synopsys: dsi: add support for non-continuous HS clock
In reply to Greg Kroah-HartmanFrom: Antonio Borneo <[email protected]>
[ Upstream commit c6d94e37bdbb6dfe7e581e937a915ab58399b8a5 ]
Current code enables the HS clock when video mode is started or to send out a HS command, and disables the HS clock to send out a LP command. This is not what DSI spec specify.
Enable HS clock either in command and in video mode. Set automatic HS clock management for panels and devices that support non-continuous HS clock.
--- drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c +7 -2
@@ -326,7 +326,6 @@ static void dw_mipi_message_config(struct dw_mipi_dsi *dsi,if (lpm) val |=CMD_MODE_ALL_LP ; -dsi_write (dsi,DSI_LPCLK_CTRL , lpm ?0 :PHY_TXREQUESTCLKHS );dsi_write (dsi,DSI_CMD_MODE_CFG , val); }
@@ -488,16 +487,22 @@ static void dw_mipi_dsi_video_mode_config(struct dw_mipi_dsi *dsi)static void dw_mipi_dsi_set_mode (struct dw_mipi_dsi *dsi,unsigned long mode_flags) { +u32 val; +dsi_write (dsi,DSI_PWR_UP ,RESET );if (mode_flags &MIPI_DSI_MODE_VIDEO ) {dsi_write (dsi,DSI_MODE_CFG ,ENABLE_VIDEO_MODE );dw_mipi_dsi_video_mode_config (dsi); -dsi_write (dsi,DSI_LPCLK_CTRL ,PHY_TXREQUESTCLKHS ); }else {dsi_write (dsi,DSI_MODE_CFG ,ENABLE_CMD_MODE ); } + val =PHY_TXREQUESTCLKHS ; +if (dsi->mode_flags &MIPI_DSI_CLOCK_NON_CONTINUOUS ) + val |=AUTO_CLKLANE_CTRL ; +dsi_write (dsi,DSI_LPCLK_CTRL , val); +dsi_write (dsi,DSI_PWR_UP ,POWERUP ); }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 76/191 arm64: topology: Stop using MPIDR for topology information
In reply to Greg Kroah-HartmanFrom: Valentin Schneider <[email protected]>
[ Upstream commit 3102bc0e6ac752cc5df896acb557d779af4d82a1 ]
In the absence of ACPI or DT topology data, we fallback to haphazardly decoding *something* out of MPIDR. Sadly, the contents of that register are mostly unusable due to the implementation leniancy and things like Aff0 having to be capped to 15 (despite being encoded on 8 bits).
Consider a simple system with a single package of 32 cores, all under the same LLC. We ought to be shoving them in the same core_sibling mask, but MPIDR is going to look like:
| CPU | 0 | ... | 15 | 16 | ... | 31 | |------+---+-----+----+----+-----+----+ | Aff0 | 0 | ... | 15 | 0 | ... | 15 | | Aff1 | 0 | ... | 0 | 1 | ... | 1 | | Aff2 | 0 | ... | 0 | 0 | ... | 0 |
Which will eventually yield
core_sibling(0-15) == 0-15
core_sibling(16-31) == 16-31
NUMA woes
=========
If we try to play games with this and set up NUMA boundaries within those groups of 16 cores via e.g. QEMU:
# Node0: 0-9; Node1: 10-19
$ qemu-system-aarch64 <blah> \
-smp 20 -numa node,cpus=0-9,nodeid=0 -numa node,cpus=10-19,nodeid=1The scheduler's MC domain (all CPUs with same LLC) is going to be built via
arch_topology.c::cpu_coregroup_mask()
In there we try to figure out a sensible mask out of the topology information we have. In short, here we'll pick the smallest of NUMA or core sibling mask.
node_mask(CPU9) == 0-9
core_sibling(CPU9) == 0-15
MC mask for CPU9 will thus be 0-9, not a problem.
node_mask(CPU10) == 10-19
core_sibling(CPU10) == 0-15
MC mask for CPU10 will thus be 10-19, not a problem.
node_mask(CPU16) == 10-19
core_sibling(CPU16) == 16-19
MC mask for CPU16 will thus be 16-19... Uh oh. CPUs 16-19 are in two different unique MC spans, and the scheduler has no idea what to make of that. That triggers the WARN_ON() added by commit
ccf74128d66c ("sched/topology: Assert non-NUMA topology masks don't (partially) overlap")
Fixing MPIDR-derived topology
=============================
We could try to come up with some cleverer scheme to figure out which of the available masks to pick, but really if one of those masks resulted from MPIDR then it should be discarded because it's bound to be bogus.
I was hoping to give MPIDR a chance for SMT, to figure out which threads are in the same core using Aff1-3 as core ID, but Sudeep and Robin pointed out to me that there are systems out there where *all* cores have non-zero values in their higher affinity fields (e.g. RK3288 has "5" in all of its cores' MPIDR.Aff1), which would expose a bogus core ID to userspace.
Stop using MPIDR for topology information. When no other source of topology information is available, mark each CPU as its own core and its NUMA node as its LLC domain.
--- arch/arm64/kernel/topology.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-)
arch/arm64/kernel/topology.c +17 -15
@@ -272,21 +272,23 @@ void store_cpu_topology(unsigned int cpuid)if (mpidr &MPIDR_UP_BITMASK )return ; -/* Create cpu topology mapping based on MPIDR. */ -if (mpidr &MPIDR_MT_BITMASK ) { -/* Multiprocessor system : Multi-threads per core */ - cpuid_topo->thread_id =MPIDR_AFFINITY_LEVEL (mpidr,0 ); - cpuid_topo->core_id =MPIDR_AFFINITY_LEVEL (mpidr,1 ); - cpuid_topo->package_id =MPIDR_AFFINITY_LEVEL (mpidr,2 ) | -MPIDR_AFFINITY_LEVEL (mpidr,3 ) <<8 ; - }else { -/* Multiprocessor system : Single-thread per core */ - cpuid_topo->thread_id =-1 ; - cpuid_topo->core_id =MPIDR_AFFINITY_LEVEL (mpidr,0 ); - cpuid_topo->package_id =MPIDR_AFFINITY_LEVEL (mpidr,1 ) | -MPIDR_AFFINITY_LEVEL (mpidr,2 ) <<8 | -MPIDR_AFFINITY_LEVEL (mpidr,3 ) <<16 ; - } +/* +* This would be the place to create cpu topology based on MPIDR. +* +* However, it cannot be trusted to depict the actual topology; some +* pieces of the architecture enforce an artificial cap on Aff0 values +* (e.g. GICv3's ICC_SGI1R_EL1 limits it to 15), leading to an +* artificial cycling of Aff1, Aff2 and Aff3 values. IOW, these end up +* having absolutely no relationship to the actual underlying system +* topology, and cannot be reasonably used as core / package ID. +* +* If the MT bit is set, Aff0 *could* be used to define a thread ID, but +* we still wouldn't be able to obtain a sane core ID. This means we +* need to entirely ignore MPIDR for any topology deduction. +*/ + cpuid_topo->thread_id =-1 ; + cpuid_topo->core_id = cpuid; + cpuid_topo->package_id =cpu_to_node (cpuid); pr_debug("CPU%u: cluster %d core %d thread %d mpidr %#016llx\n", cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 77/191 printk: reduce LOG_BUF_SHIFT range for H8300
In reply to Greg Kroah-HartmanFrom: John Ogness <[email protected]>
[ Upstream commit 550c10d28d21bd82a8bb48debbb27e6ed53262f6 ]
The .bss section for the h8300 is relatively small. A value of CONFIG_LOG_BUF_SHIFT that is larger than 19 will create a static printk ringbuffer that is too large. Limit the range appropriately for the H8300.
--- init/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
init/Kconfig +2 -1
@@ -535,7 +535,8 @@ config IKCONFIG_PROC config LOG_BUF_SHIFT int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" - range 12 25 + range 12 25 if !H8300 + range 12 19 if H8300 default 17 depends on PRINTK help
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 78/191 ia64: kprobes: Use generic kretprobe trampoline handler
In reply to Greg Kroah-HartmanFrom: Masami Hiramatsu <[email protected]>
[ Upstream commit e792ff804f49720ce003b3e4c618b5d996256a18 ]
Use the generic kretprobe trampoline handler. Don't use framepointer verification.
--- arch/ia64/kernel/kprobes.c | 77 +------------------------------------- 1 file changed, 2 insertions(+), 75 deletions(-)
arch/ia64/kernel/kprobes.c +2 -75
@@ -409,83 +409,9 @@ static void kretprobe_trampoline(void) { } -/* -* At this point the target function has been tricked into -* returning into our trampoline. Lookup the associated instance -* and then: -* - call the handler function -* - cleanup by marking the instance as unused -* - long jump back to the original return address -*/ int __kprobes trampoline_probe_handler (struct kprobe *p,struct pt_regs *regs) { -struct kretprobe_instance *ri =NULL ; -struct hlist_head *head, empty_rp; -struct hlist_node *tmp; -unsigned long flags, orig_ret_address =0 ; -unsigned long trampoline_address = - ((struct fnptr *)kretprobe_trampoline)->ip; - -INIT_HLIST_HEAD (&empty_rp); -kretprobe_hash_lock (current, &head, &flags); - -/* -* It is possible to have multiple instances associated with a given -* task either because an multiple functions in the call path -* have a return probe installed on them, and/or more than one return -* return probe was registered for a target function. -* -* We can handle this because: -* - instances are always inserted at the head of the list -* - when multiple return probes are registered for the same -* function, the first instance's ret_addr will point to the -* real return address, and all the rest will point to -* kretprobe_trampoline -*/ -hlist_for_each_entry_safe (ri, tmp, head, hlist) { -if (ri->task != current) -/* another task is sharing our hash bucket */ -continue ; - - orig_ret_address = (unsigned long )ri->ret_addr; -if (orig_ret_address != trampoline_address) -/* -* This is the real return address. Any other -* instances associated with this task are for -* other calls deeper on the call stack -*/ -break ; - } - - regs->cr_iip = orig_ret_address; - -hlist_for_each_entry_safe (ri, tmp, head, hlist) { -if (ri->task != current) -/* another task is sharing our hash bucket */ -continue ; - -if (ri->rp && ri->rp->handler) - ri->rp->handler (ri, regs); - - orig_ret_address = (unsigned long )ri->ret_addr; -recycle_rp_inst (ri, &empty_rp); - -if (orig_ret_address != trampoline_address) -/* -* This is the real return address. Any other -* instances associated with this task are for -* other calls deeper on the call stack -*/ -break ; - } -kretprobe_assert (ri, orig_ret_address, trampoline_address); - -kretprobe_hash_unlock (current, &flags); - -hlist_for_each_entry_safe (ri, tmp, &empty_rp, hlist) { -hlist_del (&ri->hlist); -kfree (ri); - } + regs->cr_iip =__kretprobe_trampoline_handler (regs, kretprobe_trampoline,NULL );/* * By returning a non-zero value, we are telling * kprobe_handler() that we don't want the post_handler
@@ -498,6 +424,7 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) { ri->ret_addr = (kprobe_opcode_t *)regs->b0; + ri->fp =NULL ;/* Replace the return addr with trampoline addr */ regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
Signature · 1 line
2.27.0
Re: [PATCH 4.19 078/191] ia64: kprobes: Use generic kretprobe trampoline handler
Hi!
Quoted 11 lines
From: Masami Hiramatsu <[email protected]>
[ Upstream commit
e792ff804f49720ce003b3e4c618b5d996256a18]Use the generic kretprobe trampoline handler. Don't use framepointer verification.
Signed-off-by Masami Hiramatsu <[email protected]>Signed-off-by Ingo Molnar <[email protected]>Signed-off-by Sasha Levin <[email protected]>
There's no explanation here why this is good idea for -stable.
Plus, this seems to depend on commit 66ada2ccae4ed4dd07ba91df3b5fdb4c11335bd1 which is not present in stable, so this will likely not build due to lack of __kretprobe_trampoline_handler symbol. (I don't have ia64 to verify):
$ grep -ri __kretprobe_trampoline_handler . ./arch/ia64/kernel/kprobes.c: regs->cr_iip = __kretprobe_trampoline_handler(regs, kretprobe_trampoline, NULL); $
I believe it should be dropped.
Best regards,
Pavel
...
Quoted 5 lines
arch/ia64/kernel/kprobes.c +1
-4-hlist_for_each_entry_safe (ri, tmp, &empty_rp, hlist) { -hlist_del (&ri->hlist); -kfree (ri); - } + regs->cr_iip =__kretprobe_trampoline_handler (regs, kretprobe_trampoline,NULL );
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Greg Kroah-Hartman
maintainerPATCH 79/191 kgdb: Make "kgdbcon" work properly with "kgdb_earlycon"
In reply to Greg Kroah-HartmanFrom: Douglas Anderson <[email protected]>
[ Upstream commit b18b099e04f450cdc77bec72acefcde7042bd1f3 ]
On my system the kernel processes the "kgdb_earlycon" parameter before the "kgdbcon" parameter. When we setup "kgdb_earlycon" we'll end up in kgdb_register_callbacks() and "kgdb_use_con" won't have been set yet so we'll never get around to starting "kgdbcon". Let's remedy this by detecting that the IO module was already registered when setting "kgdb_use_con" and registering the console then.
As part of this, to avoid pre-declaring things, move the handling of the "kgdbcon" further down in the file.
--- kernel/debug/debug_core.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-)
kernel/debug/debug_core.c +14 -8
@@ -95,14 +95,6 @@ int dbg_switch_cpu;/* Use kdb or gdbserver mode */ int dbg_kdb_mode =1 ; -static int __init opt_kgdb_con (char *str) -{ - kgdb_use_con =1 ; -return 0 ; -} - -early_param ("kgdbcon" , opt_kgdb_con); -module_param (kgdb_use_con, int,0644 );module_param (kgdbreboot, int,0644 );
@@ -820,6 +812,20 @@ static struct console kgdbcons = { .index = -1, }; +static int __init opt_kgdb_con (char *str) +{ + kgdb_use_con =1 ; + +if (kgdb_io_module_registered && !kgdb_con_registered) { +register_console (&kgdbcons); + kgdb_con_registered =1 ; + } + +return 0 ; +} + +early_param ("kgdbcon" , opt_kgdb_con); +#ifdef CONFIG_MAGIC_SYSRQ static void sysrq_handle_dbg (int key) {
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 80/191 media: uvcvideo: Fix dereference of out-of-bound list iterator
In reply to Greg Kroah-HartmanFrom: Daniel W. S. Almeida <[email protected]>
[ Upstream commit f875bcc375c738bf2f599ff2e1c5b918dbd07c45 ]
Fixes the following coccinelle report:
drivers/media/usb/uvc/uvc_ctrl.c:1860:5-11:
ERROR: invalid reference to the index variable of the iterator on line 1854
by adding a boolean variable to check if the loop has found the
Found using - Coccinelle (http://coccinelle.lip6.fr)
[Replace cursor variable with bool found]
--- drivers/media/usb/uvc/uvc_ctrl.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/media/usb/uvc/uvc_ctrl.c +9 -4
@@ -1853,30 +1853,35 @@ int uvc_xu_ctrl_query(struct uvc_video_chain *chain, {struct uvc_entity *entity;struct uvc_control *ctrl; -unsigned int i, found =0 ; +unsigned int i; +bool found;u32 reqflags;u16 size;u8 *data =NULL ;int ret;/* Find the extension unit. */ + found = false;list_for_each_entry (entity, &chain->entities, chain) {if (UVC_ENTITY_TYPE (entity) ==UVC_VC_EXTENSION_UNIT && - entity->id == xqry->unit) + entity->id == xqry->unit) { + found = true;break ; + } } -if (entity->id != xqry->unit) { +if (!found) {uvc_trace (UVC_TRACE_CONTROL ,"Extension unit %u not found.\n" , xqry->unit);return -ENOENT ; }/* Find the control and perform delayed initialization if needed. */ + found = false;for (i =0 ; i < entity->ncontrols; ++i) { ctrl = &entity->controls[i];if (ctrl->index == xqry->selector -1 ) { - found =1 ; + found = true;break ; } }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 81/191 riscv: Define AT_VECTOR_SIZE_ARCH for ARCH_DLINFO
In reply to Greg Kroah-HartmanFrom: Zong Li <[email protected]>
[ Upstream commit b5fca7c55f9fbab5ad732c3bce00f31af6ba5cfa ]
AT_VECTOR_SIZE_ARCH should be defined with the maximum number of NEW_AUX_ENT entries that ARCH_DLINFO can contain, but it wasn't defined for RISC-V at all even though ARCH_DLINFO will contain one NEW_AUX_ENT for the VDSO address.
--- arch/riscv/include/uapi/asm/auxvec.h | 3 +++ 1 file changed, 3 insertions(+)
arch/riscv/include/uapi/asm/auxvec.h +3 -0
@@ -21,4 +21,7 @@/* vDSO location */ #define AT_SYSINFO_EHDR 33 +/* entries in ARCH_DLINFO */ +#define AT_VECTOR_SIZE_ARCH 1 +#endif /* _UAPI_ASM_RISCV_AUXVEC_H */
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 82/191 cpufreq: sti-cpufreq: add stih418 support
In reply to Greg Kroah-HartmanFrom: Alain Volmat <[email protected]>
[ Upstream commit 01a163c52039e9426c7d3d3ab16ca261ad622597 ]
The STiH418 can be controlled the same way as STiH407 & STiH410 regarding cpufreq.
--- drivers/cpufreq/sti-cpufreq.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/cpufreq/sti-cpufreq.c +4 -2
@@ -144,7 +144,8 @@ static const struct reg_field sti_stih407_dvfs_regfields[DVFS_MAX_REGFIELDS] = {static const struct reg_field *sti_cpufreq_match (void ) {if (of_machine_is_compatible ("st,stih407" ) || -of_machine_is_compatible ("st,stih410" )) +of_machine_is_compatible ("st,stih410" ) || +of_machine_is_compatible ("st,stih418" ))return sti_stih407_dvfs_regfields;return NULL ;
@@ -261,7 +262,8 @@ static int sti_cpufreq_init(void)int ret;if ((!of_machine_is_compatible ("st,stih407" )) && - (!of_machine_is_compatible ("st,stih410" ))) + (!of_machine_is_compatible ("st,stih410" )) && + (!of_machine_is_compatible ("st,stih418" )))return -ENODEV ; ddata.cpu =get_cpu_device (0 );
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 83/191 USB: adutux: fix debugging
In reply to Greg Kroah-HartmanFrom: Oliver Neukum <[email protected]>
[ Upstream commit c56150c1bc8da5524831b1dac2eec3c67b89f587 ]
Handling for removal of the controller was missing at one place. Add it.
--- drivers/usb/misc/adutux.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/misc/adutux.c +1 -0
@@ -209,6 +209,7 @@ static void adu_interrupt_out_callback(struct urb *urb)if (status !=0 ) {if ((status != -ENOENT ) && + (status != -ESHUTDOWN ) && (status != -ECONNRESET )) { dev_dbg(&dev->udev->dev, "%s :nonzero status received: %d\n", __func__,
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 84/191 uio: free uio id after uio file node is freed
In reply to Greg Kroah-HartmanFrom: Lang Dai <[email protected]>
[ Upstream commit 8fd0e2a6df262539eaa28b0a2364cca10d1dc662 ]
uio_register_device() do two things.
1) get an uio id from a global pool, e.g. the id is <A>
2) create file nodes like /sys/class/uio/uio<A>
uio_unregister_device() do two things.
1) free the uio id <A> and return it to the global pool
2) free the file node /sys/class/uio/uio<A>
There is a situation is that one worker is calling uio_unregister_device(), and another worker is calling uio_register_device().
If the two workers are X and Y, they go as below sequence,
1) X free the uio id <AAA>
2) Y get an uio id <AAA>
3) Y create file node /sys/class/uio/uio<AAA>
4) X free the file note /sys/class/uio/uio<AAA>
Then it will failed at the 3rd step and cause the phenomenon we saw as it is creating a duplicated file node.
Failure reports as follows:
sysfs: cannot create duplicate filename '/class/uio/uio10'
Call Trace: sysfs_do_create_link_sd.isra.2+0x9e/0xb0 sysfs_create_link+0x25/0x40 device_add+0x2c4/0x640 __uio_register_device+0x1c5/0x576 [uio] adf_uio_init_bundle_dev+0x231/0x280 [intel_qat] adf_uio_register+0x1c0/0x340 [intel_qat] adf_dev_start+0x202/0x370 [intel_qat] adf_dev_start_async+0x40/0xa0 [intel_qat] process_one_work+0x14d/0x410 worker_thread+0x4b/0x460 kthread+0x105/0x140 ? process_one_work+0x410/0x410 ? kthread_bind+0x40/0x40 ret_from_fork+0x1f/0x40 Code: 85 c0 48 89 c3 74 12 b9 00 10 00 00 48 89 c2 31 f6 4c 89 ef e8 ec c4 ff ff 4c 89 e2 48 89 de 48 c7 c7 e8 b4 ee b4 e8 6a d4 d7 ff <0f> 0b 48 89 df e8 20 fa f3 ff 5b 41 5c 41 5d 5d c3 66 0f 1f 84 ---[ end trace a7531c1ed5269e84 ]--- c6xxvf b002:00:00.0: Failed to register UIO devices c6xxvf b002:00:00.0: Failed to register UIO devices
--- drivers/uio/uio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/uio/uio.c +2 -2
@@ -1008,8 +1008,6 @@ void uio_unregister_device(struct uio_info *info) idev = info->uio_dev; -uio_free_minor (idev); -mutex_lock (&idev->info_lock);uio_dev_del_attributes (idev);
@@ -1021,6 +1019,8 @@ void uio_unregister_device(struct uio_info *info)device_unregister (&idev->dev); +uio_free_minor (idev); +return ; }EXPORT_SYMBOL_GPL (uio_unregister_device);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 85/191 usb: xhci: omit duplicate actions when suspending a runtime suspended host.
In reply to Greg Kroah-HartmanFrom: Peter Chen <[email protected]>
[ Upstream commit 18a367e8947d72dd91b6fc401e88a2952c6363f7 ]
If the xhci-plat.c is the platform driver, after the runtime pm is enabled, the xhci_suspend is called if nothing is connected on the port. When the system goes to suspend, it will call xhci_suspend again if USB wakeup is enabled.
Since the runtime suspend wakeup setting is not always the same as system suspend wakeup setting, eg, at runtime suspend we always need wakeup if the controller is in low power mode; but at system suspend, we may not need wakeup. So, we move the judgement after changing wakeup setting.
[commit message rewording -Mathias]
--- drivers/usb/host/xhci.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/host/xhci.c +5 -2
@@ -972,12 +972,15 @@ int xhci_suspend(struct xhci_hcd *xhci, bool do_wakeup) xhci->shared_hcd->state != HC_STATE_SUSPENDED)return -EINVAL ; -xhci_dbc_suspend (xhci); -/* Clear root port wake on bits if wakeup not allowed. */ if (!do_wakeup)xhci_disable_port_wake_on_bits (xhci); +if (!HCD_HW_ACCESSIBLE (hcd)) +return 0 ; + +xhci_dbc_suspend (xhci); +/* Don't poll the roothubs on bus suspend. */ xhci_dbg (xhci,"%s: stopping port polling.\n" , __func__);clear_bit (HCD_FLAG_POLL_RH , &hcd->flags);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 86/191 arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE
In reply to Greg Kroah-HartmanFrom: Zhengyuan Liu <[email protected]>
[ Upstream commit a194c5f2d2b3a05428805146afcabe5140b5d378 ]
The @node passed to cpumask_of_node() can be NUMA_NO_NODE, in that case it will trigger the following WARN_ON(node >= nr_node_ids) due to mismatched data types of @node and @nr_node_ids. Actually we should return cpu_all_mask just like most other architectures do if passed NUMA_NO_NODE.
Also add a similar check to the inline cpumask_of_node() in numa.h.
--- arch/arm64/include/asm/numa.h | 3 +++ arch/arm64/mm/numa.c | 6 +++++- 2 files changed, 8 insertions(+), 1 deletion(-)
arch/arm64/include/asm/numa.h +3 -0
@@ -25,6 +25,9 @@ const struct cpumask *cpumask_of_node(int node);/* Returns a pointer to the cpumask of CPUs on Node 'node'. */ static inline const struct cpumask *cpumask_of_node (int node) { +if (node ==NUMA_NO_NODE ) +return cpu_all_mask; +return node_to_cpumask_map[node]; }#endif
arch/arm64/mm/numa.c +5 -1
@@ -58,7 +58,11 @@ EXPORT_SYMBOL(node_to_cpumask_map);*/ const struct cpumask *cpumask_of_node (int node) { -if (WARN_ON (node >= nr_node_ids)) + +if (node ==NUMA_NO_NODE ) +return cpu_all_mask; + +if (WARN_ON (node <0 || node >= nr_node_ids))return cpu_none_mask;if (WARN_ON (node_to_cpumask_map[node] ==NULL ))
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 87/191 xfs: dont free rt blocks when were doing a REMAP bunmapi call
In reply to Greg Kroah-HartmanFrom: Darrick J. Wong <[email protected]>
[ Upstream commit 8df0fa39bdd86ca81a8d706a6ed9d33cc65ca625 ]
When callers pass XFS_BMAPI_REMAP into xfs_bunmapi, they want the extent to be unmapped from the given file fork without the extent being freed. We do this for non-rt files, but we forgot to do this for realtime files. So far this isn't a big deal since nobody makes a bunmapi call to a rt file with the REMAP flag set, but don't leave a logic bomb.
--- fs/xfs/libxfs/xfs_bmap.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)
fs/xfs/libxfs/xfs_bmap.c +12 -7
@@ -4920,20 +4920,25 @@ xfs_bmap_del_extent_real( flags =XFS_ILOG_CORE ;if (whichfork ==XFS_DATA_FORK &&XFS_IS_REALTIME_INODE (ip)) { -xfs_fsblock_t bno;xfs_filblks_t len;xfs_extlen_t mod; - bno =div_u64_rem (del->br_startblock, mp->m_sb.sb_rextsize, - &mod); -ASSERT (mod ==0 ); len =div_u64_rem (del->br_blockcount, mp->m_sb.sb_rextsize, &mod);ASSERT (mod ==0 ); - error =xfs_rtfree_extent (tp, bno, (xfs_extlen_t )len); -if (error) - goto done; +if (!(bflags &XFS_BMAPI_REMAP )) { +xfs_fsblock_t bno; + + bno =div_u64_rem (del->br_startblock, + mp->m_sb.sb_rextsize, &mod); +ASSERT (mod ==0 ); + + error =xfs_rtfree_extent (tp, bno, (xfs_extlen_t )len); +if (error) + goto done; + } + do_fx =0 ; nblks = len * mp->m_sb.sb_rextsize; qfield =XFS_TRANS_DQ_RTBCOUNT ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 88/191 ACPI: Add out of bounds and numa_off protections to pxm_to_node()
In reply to Greg Kroah-HartmanFrom: Jonathan Cameron <[email protected]>
[ Upstream commit 8a3decac087aa897df5af04358c2089e52e70ac4 ]
The function should check the validity of the pxm value before using it to index the pxm_to_node_map[] array.
Whilst hardening this code may be good in general, the main intent here is to enable following patches that use this function to replace acpi_map_pxm_to_node() for non SRAT usecases which should return NO_NUMA_NODE for PXM entries not matching with those in SRAT.
--- drivers/acpi/numa.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/acpi/numa.c +1 -1
@@ -46,7 +46,7 @@ int acpi_numa __initdata;int pxm_to_node (int pxm) { -if (pxm <0 ) +if (pxm <0 || pxm >=MAX_PXM_DOMAINS || numa_off)return NUMA_NO_NODE ;return pxm_to_node_map[pxm]; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 89/191 drivers/net/wan/hdlc_fr: Correctly handle special skb->protocol values
In reply to Greg Kroah-HartmanFrom: Xie He <[email protected]>
[ Upstream commit 8306266c1d51aac9aa7aa907fe99032a58c6382c ]
The fr_hard_header function is used to prepend the header to skbs before transmission. It is used in 3 situations:
1) When a control packet is generated internally in this driver;
2) When a user sends an skb on an Ethernet-emulating PVC device;
3) When a user sends an skb on a normal PVC device.
These 3 situations need to be handled differently by fr_hard_header. Different headers should be prepended to the skb in different situations.
Currently fr_hard_header distinguishes these 3 situations using skb->protocol. For situation 1 and 2, a special skb->protocol value will be assigned before calling fr_hard_header, so that it can recognize these 2 situations. All skb->protocol values other than these special ones are treated by fr_hard_header as situation 3.
However, it is possible that in situation 3, the user sends an skb with one of the special skb->protocol values. In this case, fr_hard_header would incorrectly treat it as situation 1 or 2.
This patch tries to solve this issue by using skb->dev instead of skb->protocol to distinguish between these 3 situations. For situation 1, skb->dev would be NULL; for situation 2, skb->dev->type would be ARPHRD_ETHER; and for situation 3, skb->dev->type would be ARPHRD_DLCI.
This way fr_hard_header would be able to distinguish these 3 situations correctly regardless what skb->protocol value the user tries to use in situation 3.
--- drivers/net/wan/hdlc_fr.c | 98 ++++++++++++++++++++------------------- 1 file changed, 51 insertions(+), 47 deletions(-)
drivers/net/wan/hdlc_fr.c +51 -47
@@ -276,63 +276,69 @@ static inline struct net_device **get_dev_p(struct pvc_device *pvc,static int fr_hard_header (struct sk_buff **skb_p,u16 dlci) { -u16 head_len;struct sk_buff *skb = *skb_p; -switch (skb->protocol) { -case cpu_to_be16 (NLPID_CCITT_ANSI_LMI ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_CCITT_ANSI_LMI ; -break ; - -case cpu_to_be16 (NLPID_CISCO_LMI ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_CISCO_LMI ; -break ; - -case cpu_to_be16 (ETH_P_IP ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_IP ; -break ; - -case cpu_to_be16 (ETH_P_IPV6 ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_IPV6 ; -break ; - -case cpu_to_be16 (ETH_P_802_3 ): - head_len =10 ; -if (skb_headroom (skb) < head_len) { -struct sk_buff *skb2 =skb_realloc_headroom (skb, - head_len); +if (!skb->dev) {/* Control packets */ +switch (dlci) { +case LMI_CCITT_ANSI_DLCI : +skb_push (skb,4 ); + skb->data[3 ] =NLPID_CCITT_ANSI_LMI ; +break ; + +case LMI_CISCO_DLCI : +skb_push (skb,4 ); + skb->data[3 ] =NLPID_CISCO_LMI ; +break ; + +default : +return -EINVAL ; + } + + }else if (skb->dev->type ==ARPHRD_DLCI ) { +switch (skb->protocol) { +case htons (ETH_P_IP ): +skb_push (skb,4 ); + skb->data[3 ] =NLPID_IP ; +break ; + +case htons (ETH_P_IPV6 ): +skb_push (skb,4 ); + skb->data[3 ] =NLPID_IPV6 ; +break ; + +default : +skb_push (skb,10 ); + skb->data[3 ] =FR_PAD ; + skb->data[4 ] =NLPID_SNAP ; +/* OUI 00-00-00 indicates an Ethertype follows */ + skb->data[5 ] =0x00 ; + skb->data[6 ] =0x00 ; + skb->data[7 ] =0x00 ; +/* This should be an Ethertype: */ + *(__be16 *)(skb->data +8 ) = skb->protocol; + } + + }else if (skb->dev->type ==ARPHRD_ETHER ) { +if (skb_headroom (skb) <10 ) { +struct sk_buff *skb2 =skb_realloc_headroom (skb,10 );if (!skb2)return -ENOBUFS ;dev_kfree_skb (skb); skb = *skb_p = skb2; } -skb_push (skb, head_len); +skb_push (skb,10 ); skb->data[3 ] =FR_PAD ; skb->data[4 ] =NLPID_SNAP ; - skb->data[5 ] =FR_PAD ; +/* OUI 00-80-C2 stands for the 802.1 organization */ + skb->data[5 ] =0x00 ; skb->data[6 ] =0x80 ; skb->data[7 ] =0xC2 ; +/* PID 00-07 stands for Ethernet frames without FCS */ skb->data[8 ] =0x00 ; - skb->data[9 ] =0x07 ;/* bridged Ethernet frame w/out FCS */ -break ; + skb->data[9 ] =0x07 ; -default : - head_len =10 ; -skb_push (skb, head_len); - skb->data[3 ] =FR_PAD ; - skb->data[4 ] =NLPID_SNAP ; - skb->data[5 ] =FR_PAD ; - skb->data[6 ] =FR_PAD ; - skb->data[7 ] =FR_PAD ; - *(__be16 *)(skb->data +8 ) = skb->protocol; + }else { +return -EINVAL ; }dlci_to_q922 (skb->data, dlci);
@@ -428,8 +434,8 @@ static netdev_tx_t pvc_xmit(struct sk_buff *skb, struct net_device *dev)skb_put (skb, pad);memset (skb->data + len,0 , pad); } - skb->protocol =cpu_to_be16 (ETH_P_802_3 ); } + skb->dev = dev;if (!fr_hard_header (&skb, pvc->dlci)) { dev->stats.tx_bytes += skb->len; dev->stats.tx_packets++;
@@ -497,10 +503,8 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)memset (skb->data,0 , len);skb_reserve (skb,4 );if (lmi ==LMI_CISCO ) { - skb->protocol =cpu_to_be16 (NLPID_CISCO_LMI );fr_hard_header (&skb,LMI_CISCO_DLCI ); }else { - skb->protocol =cpu_to_be16 (NLPID_CCITT_ANSI_LMI );fr_hard_header (&skb,LMI_CCITT_ANSI_DLCI ); } data =skb_tail_pointer (skb);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 90/191 bus/fsl_mc: Do not rely on caller to provide non NULL mc_io
In reply to Greg Kroah-HartmanFrom: Diana Craciun <[email protected]>
[ Upstream commit 5026cf605143e764e1785bbf9158559d17f8d260 ]
Before destroying the mc_io, check first that it was allocated.
--- drivers/bus/fsl-mc/mc-io.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/bus/fsl-mc/mc-io.c +6 -1
@@ -129,7 +129,12 @@ error_destroy_mc_io:*/ void fsl_destroy_mc_io (struct fsl_mc_io *mc_io) { -struct fsl_mc_device *dpmcp_dev = mc_io->dpmcp_dev; +struct fsl_mc_device *dpmcp_dev; + +if (!mc_io) +return ; + + dpmcp_dev = mc_io->dpmcp_dev;if (dpmcp_dev)fsl_mc_io_unset_dpmcp (mc_io);
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 91/191 power: supply: test_power: add missing newlines when printing parameters by sysfs
In reply to Greg Kroah-HartmanFrom: Xiongfeng Wang <[email protected]>
[ Upstream commit c07fa6c1631333f02750cf59f22b615d768b4d8f ]
When I cat some module parameters by sysfs, it displays as follows. It's better to add a newline for easy reading.
root@syzkaller:~# cd /sys/module/test_power/parameters/ root@syzkaller:/sys/module/test_power/parameters# cat ac_online onroot@syzkaller:/sys/module/test_power/parameters# cat battery_present trueroot@syzkaller:/sys/module/test_power/parameters# cat battery_health goodroot@syzkaller:/sys/module/test_power/parameters# cat battery_status dischargingroot@syzkaller:/sys/module/test_power/parameters# cat battery_technology LIONroot@syzkaller:/sys/module/test_power/parameters# cat usb_online onroot@syzkaller:/sys/module/test_power/parameters#
--- drivers/power/supply/test_power.c | 6 ++++++ 1 file changed, 6 insertions(+)
drivers/power/supply/test_power.c +6 -0
@@ -344,6 +344,7 @@ static int param_set_ac_online(const char *key, const struct kernel_param *kp)static int param_get_ac_online (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_ac_online, ac_online,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -357,6 +358,7 @@ static int param_set_usb_online(const char *key, const struct kernel_param *kp)static int param_get_usb_online (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_ac_online, usb_online,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -371,6 +373,7 @@ static int param_set_battery_status(const char *key,static int param_get_battery_status (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_status, battery_status,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -385,6 +388,7 @@ static int param_set_battery_health(const char *key,static int param_get_battery_health (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_health, battery_health,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -400,6 +404,7 @@ static int param_get_battery_present(char *buffer, const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_present, battery_present,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -417,6 +422,7 @@ static int param_get_battery_technology(char *buffer, {strcpy (buffer,map_get_key (map_technology, battery_technology,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 92/191 drm/amd/display: HDMI remote sink need mode validation for Linux
In reply to Greg Kroah-HartmanFrom: Fangzhi Zuo <[email protected]>
[ Upstream commit 95d620adb48f7728e67d82f56f756e8d451cf8d2 ]
[Why]
Currently mode validation is bypassed if remote sink exists. That leads to mode set issue when a BW bottle neck exists in the link path, e.g., a DP-to-HDMI converter that only supports HDMI 1.4.
Any invalid mode passed to Linux user space will cause the modeset failure due to limitation of Linux user space implementation.
[How]
Mode validation is skipped only if in edid override. For real remote sink, clock limit check should be done for HDMI remote sink.
Have HDMI related remote sink going through mode validation to elimiate modes which pixel clock exceeds BW limitation.
--- drivers/gpu/drm/amd/display/dc/core/dc_link.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/display/dc/core/dc_link.c +1 -1
@@ -2009,7 +2009,7 @@ enum dc_status dc_link_validate_mode_timing(/* A hack to avoid failing any modes for EDID override feature on * topology change such as lower quality cable for DP or different dongle */ -if (link->remote_sinks[0 ]) +if (link->remote_sinks[0 ] && link->remote_sinks[0 ]->sink_signal ==SIGNAL_TYPE_VIRTUAL )return DC_OK ;/* Passive Dongle */
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 93/191 btrfs: fix replace of seed device
In reply to Greg Kroah-HartmanFrom: Anand Jain <[email protected]>
[ Upstream commit c6a5d954950c5031444173ad2195efc163afcac9 ]
If you replace a seed device in a sprouted fs, it appears to have successfully replaced the seed device, but if you look closely, it didn't. Here is an example.
$ mkfs.btrfs /dev/sda $ btrfstune -S1 /dev/sda $ mount /dev/sda /btrfs $ btrfs device add /dev/sdb /btrfs $ umount /btrfs $ btrfs device scan --forget $ mount -o device=/dev/sda /dev/sdb /btrfs $ btrfs replace start -f /dev/sda /dev/sdc /btrfs $ echo $? 0
BTRFS info (device sdb): dev_replace from /dev/sda (devid 1) to /dev/sdc started BTRFS info (device sdb): dev_replace from /dev/sda (devid 1) to /dev/sdc finished
$ btrfs fi show Label: none uuid: ab2c88b7-be81-4a7e-9849-c3666e7f9f4f Total devices 2 FS bytes used 256.00KiB devid 1 size 3.00GiB used 520.00MiB path /dev/sdc devid 2 size 3.00GiB used 896.00MiB path /dev/sdb
Label: none uuid: 10bd3202-0415-43af-96a8-d5409f310a7e
Total devices 1 FS bytes used 128.00KiB
devid 1 size 3.00GiB used 536.00MiB path /dev/sda
So as per the replace start command and kernel log replace was successful. Now let's try to clean mount.
$ umount /btrfs $ btrfs device scan --forget
$ mount -o device=/dev/sdc /dev/sdb /btrfs mount: /btrfs: wrong fs type, bad option, bad superblock on /dev/sdb, missing codepage or helper program, or other error.
[ 636.157517] BTRFS error (device sdc): failed to read chunk tree: -2 [ 636.180177] BTRFS error (device sdc): open_ctree failed
That's because per dev items it is still looking for the original seed device.
$ btrfs inspect-internal dump-tree -d /dev/sdb
item 0 key (DEV_ITEMS DEV_ITEM 1) itemoff 16185 itemsize 98 devid 1 total_bytes 3221225472 bytes_used 545259520 io_align 4096 io_width 4096 sector_size 4096 type 0 generation 6 start_offset 0 dev_group 0 seek_speed 0 bandwidth 0 uuid 59368f50-9af2-4b17-91da-8a783cc418d4 <--- seed uuid fsid 10bd3202-0415-43af-96a8-d5409f310a7e <--- seed fsid item 1 key (DEV_ITEMS DEV_ITEM 2) itemoff 16087 itemsize 98 devid 2 total_bytes 3221225472 bytes_used 939524096 io_align 4096 io_width 4096 sector_size 4096 type 0 generation 0 start_offset 0 dev_group 0 seek_speed 0 bandwidth 0 uuid 56a0a6bc-4630-4998-8daf-3c3030c4256a <- sprout uuid fsid ab2c88b7-be81-4a7e-9849-c3666e7f9f4f <- sprout fsid
But the replaced target has the following uuid+fsid in its superblock which doesn't match with the expected uuid+fsid in its devitem.
$ btrfs in dump-super /dev/sdc | egrep '^generation|dev_item.uuid|dev_item.fsid|devid' generation 20 dev_item.uuid 59368f50-9af2-4b17-91da-8a783cc418d4 dev_item.fsid ab2c88b7-be81-4a7e-9849-c3666e7f9f4f [match] dev_item.devid 1
So if you provide the original seed device the mount shall be successful. Which so long happening in the test case btrfs/163.
$ btrfs device scan --forget $ mount -o device=/dev/sda /dev/sdb /btrfs
Fix in this patch:
If a seed is not sprouted then there is no replacement of it, because of its read-only filesystem with a read-only device. Similarly, in the case of a sprouted filesystem, the seed device is still read only. So, mark it as you can't replace a seed device, you can only add a new device and then delete the seed device. If replace is attempted then returns -EINVAL.
--- fs/btrfs/dev-replace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/dev-replace.c +1 -1
@@ -190,7 +190,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,int ret =0 ; *device_out =NULL ; -if (fs_info->fs_devices->seeding) { +if (srcdev->fs_devices->seeding) {btrfs_err (fs_info,"the filesystem is a seed filesystem!" );return -EINVAL ; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 94/191 md/bitmap: md_bitmap_get_counter returns wrong blocks
In reply to Greg Kroah-HartmanFrom: Zhao Heming <[email protected]>
[ Upstream commit d837f7277f56e70d82b3a4a037d744854e62f387 ]
md_bitmap_get_counter() has code:
```
if (bitmap->bp[page].hijacked ||
bitmap->bp[page].map == NULL)
csize = ((sector_t)1) << (bitmap->chunkshift +
PAGE_COUNTER_SHIFT - 1);
```The minus 1 is wrong, this branch should report 2048 bits of space. With "-1" action, this only report 1024 bit of space.
This bug code returns wrong blocks, but it doesn't inflence bitmap logic:
1. Most callers focus this function return value (the counter of offset), not the parameter blocks.
2. The bug is only triggered when hijacked is true or map is NULL. the hijacked true condition is very rare.
the "map == null" only true when array is creating or resizing.
3. Even the caller gets wrong blocks, current code makes caller just to call md_bitmap_get_counter() one more time.
--- drivers/md/md-bitmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/md-bitmap.c +1 -1
@@ -1371,7 +1371,7 @@ __acquires(bitmap->lock)if (bitmap->bp[page].hijacked || bitmap->bp[page].map ==NULL ) csize = ((sector_t )1 ) << (bitmap->chunkshift + -PAGE_COUNTER_SHIFT -1 ); +PAGE_COUNTER_SHIFT );else csize = ((sector_t )1 ) << bitmap->chunkshift; *blocks = csize - (offset & (csize -1 ));
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 95/191 bnxt_en: Log unknown link speed appropriately.
In reply to Greg Kroah-HartmanFrom: Michael Chan <[email protected]>
[ Upstream commit 8eddb3e7ce124dd6375d3664f1aae13873318b0f ]
If the VF virtual link is set to always enabled, the speed may be unknown when the physical link is down. The driver currently logs the link speed as 4294967295 Mbps which is SPEED_UNKNOWN. Modify the link up log message as "speed unknown" which makes more sense.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +5 -1
@@ -6326,6 +6326,11 @@ static void bnxt_report_link(struct bnxt *bp)u16 fec;netif_carrier_on (bp->dev); + speed =bnxt_fw_to_ethtool_speed (bp->link_info.link_speed); +if (speed ==SPEED_UNKNOWN ) { +netdev_info (bp->dev,"NIC Link is Up, speed unknown\n" ); +return ; + }if (bp->link_info.duplex ==BNXT_LINK_DUPLEX_FULL ) duplex ="full" ;else
@@ -6338,7 +6343,6 @@ static void bnxt_report_link(struct bnxt *bp) flow_ctrl ="ON - receive" ; else flow_ctrl ="none" ; - speed =bnxt_fw_to_ethtool_speed (bp->link_info.link_speed);netdev_info (bp->dev,"NIC Link is Up, %u Mbps %s duplex, Flow control: %s\n" , speed, duplex, flow_ctrl);if (bp->flags &BNXT_FLAG_EEE_CAP )
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 96/191 rpmsg: glink: Use complete_all for open states
In reply to Greg Kroah-HartmanFrom: Chris Lew <[email protected]>
[ Upstream commit 4fcdaf6e28d11e2f3820d54dd23cd12a47ddd44e ]
The open_req and open_ack completion variables are the state variables to represet a remote channel as open. Use complete_all so there are no races with waiters and using completion_done.
--- drivers/rpmsg/qcom_glink_native.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/rpmsg/qcom_glink_native.c +3 -3
@@ -970,7 +970,7 @@ static int qcom_glink_rx_open_ack(struct qcom_glink *glink, unsigned int lcid)return -EINVAL ; } -complete (&channel->open_ack); +complete_all (&channel->open_ack);return 0 ; }
@@ -1178,7 +1178,7 @@ static int qcom_glink_announce_create(struct rpmsg_device *rpdev)__be32 *val = defaults;int size; -if (glink->intentless) +if (glink->intentless || !completion_done (&channel->open_ack))return 0 ; prop =of_find_property (np,"qcom,intents" ,NULL );
@@ -1413,7 +1413,7 @@ static int qcom_glink_rx_open(struct qcom_glink *glink, unsigned int rcid, channel->rcid = ret;spin_unlock_irqrestore (&glink->idr_lock, flags); -complete (&channel->open_req); +complete_all (&channel->open_req);if (create_device) { rpdev =kzalloc (sizeof (*rpdev),GFP_KERNEL );
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 97/191 clk: ti: clockdomain: fix static checker warning
In reply to Greg Kroah-HartmanFrom: Tero Kristo <[email protected]>
[ Upstream commit b7a7943fe291b983b104bcbd2f16e8e896f56590 ]
Fix a memory leak induced by not calling clk_put after doing of_clk_get.
--- drivers/clk/ti/clockdomain.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/clk/ti/clockdomain.c +2 -0
@@ -146,10 +146,12 @@ static void __init of_ti_clockdomain_setup(struct device_node *node)if (clk_hw_get_flags (clk_hw) &CLK_IS_BASIC ) {pr_warn ("can't setup clkdm for basic clk %s\n" ,__clk_get_name (clk)); +clk_put (clk);continue ; }to_clk_hw_omap (clk_hw)->clkdm_name = clkdm_name;omap2_init_clk_clkdm (clk_hw); +clk_put (clk); } }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 98/191 net: 9p: initialize sun_server.sun_path to have addrs value only when addr is valid
In reply to Greg Kroah-HartmanFrom: Anant Thazhemadam <[email protected]>
[ Upstream commit 7ca1db21ef8e0e6725b4d25deed1ca196f7efb28 ]
In p9_fd_create_unix, checking is performed to see if the addr (passed as an argument) is NULL or not.
However, no check is performed to see if addr is a valid address, i.e., it doesn't entirely consist of only 0's.
The initialization of sun_server.sun_path to be equal to this faulty addr value leads to an uninitialized variable, as detected by KMSAN. Checking for this (faulty addr) and returning a negative error number appropriately, resolves this issue.
--- net/9p/trans_fd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/9p/trans_fd.c +1 -1
@@ -1038,7 +1038,7 @@ p9_fd_create_unix(struct p9_client *client, const char *addr, char *args) csocket =NULL ; -if (addr ==NULL ) +if (!addr || !strlen (addr))return -EINVAL ;if (strlen (addr) >=UNIX_PATH_MAX ) {
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 99/191 drivers: watchdog: rdc321x_wdt: Fix race condition bugs
In reply to Greg Kroah-HartmanFrom: Madhuparna Bhowmik <[email protected]>
[ Upstream commit 4b2e7f99cdd314263c9d172bc17193b8b6bba463 ]
In rdc321x_wdt_probe(), rdc321x_wdt_device.queue is initialized after misc_register(), hence if ioctl is called before its initialization which can call rdc321x_wdt_start() function, it will see an uninitialized value of rdc321x_wdt_device.queue, hence initialize it before misc_register().
Also, rdc321x_wdt_device.default_ticks is accessed in reset() function called from write callback, thus initialize it before misc_register().
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/watchdog/rdc321x_wdt.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/watchdog/rdc321x_wdt.c +2 -3
@@ -244,6 +244,8 @@ static int rdc321x_wdt_probe(struct platform_device *pdev) rdc321x_wdt_device.sb_pdev = pdata->sb_pdev; rdc321x_wdt_device.base_reg = r->start; + rdc321x_wdt_device.queue =0 ; + rdc321x_wdt_device.default_ticks = ticks; err =misc_register (&rdc321x_wdt_misc);if (err <0 ) {
@@ -258,14 +260,11 @@ static int rdc321x_wdt_probe(struct platform_device *pdev) rdc321x_wdt_device.base_reg, RDC_WDT_RST);init_completion (&rdc321x_wdt_device.stop); - rdc321x_wdt_device.queue =0 ;clear_bit (0 , &rdc321x_wdt_device.inuse);timer_setup (&rdc321x_wdt_device.timer, rdc321x_wdt_trigger,0 ); - rdc321x_wdt_device.default_ticks = ticks; -dev_info (&pdev->dev,"watchdog init success\n" );return 0 ;
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 100/191 ext4: Detect already used quota file early
In reply to Greg Kroah-HartmanFrom: Jan Kara <[email protected]>
[ Upstream commit e0770e91424f694b461141cbc99adf6b23006b60 ]
When we try to use file already used as a quota file again (for the same or different quota type), strange things can happen. At the very least lockdep annotations may be wrong but also inode flags may be wrongly set / reset. When the file is used for two quota types at once we can even corrupt the file and likely crash the kernel. Catch all these cases by checking whether passed file is already used as quota file and bail early in that case.
This fixes occasional generic/219 failure due to lockdep complaint.
--- fs/ext4/super.c | 5 +++++ 1 file changed, 5 insertions(+)
fs/ext4/super.c +5 -0
@@ -5752,6 +5752,11 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,/* Quotafile not on the same filesystem? */ if (path->dentry->d_sb != sb)return -EXDEV ; + +/* Quota already enabled for this file? */ +if (IS_NOQUOTA (d_inode (path->dentry))) +return -EBUSY ; +/* Journaling quota? */ if (EXT4_SB (sb)->s_qf_names[type]) {/* Quotafile not in fs root? */
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 101/191 gfs2: add validation checks for size of superblock
In reply to Greg Kroah-HartmanFrom: Anant Thazhemadam <[email protected]>
[ Upstream commit 0ddc5154b24c96f20e94d653b0a814438de6032b ]
In gfs2_check_sb(), no validation checks are performed with regards to the size of the superblock.
syzkaller detected a slab-out-of-bounds bug that was primarily caused because the block size for a superblock was set to zero.
A valid size for a superblock is a power of 2 between 512 and PAGE_SIZE. Performing validation checks and ensuring that the size of the superblock is valid fixes this bug.
--- fs/gfs2/ops_fstype.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)
fs/gfs2/ops_fstype.c +11 -7
@@ -161,15 +161,19 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)return -EINVAL ; } -/* If format numbers match exactly, we're done. */ - -if (sb->sb_fs_format ==GFS2_FORMAT_FS && - sb->sb_multihost_format ==GFS2_FORMAT_MULTI ) -return 0 ; +if (sb->sb_fs_format !=GFS2_FORMAT_FS || + sb->sb_multihost_format !=GFS2_FORMAT_MULTI ) { +fs_warn (sdp,"Unknown on-disk format, unable to mount\n" ); +return -EINVAL ; + } -fs_warn (sdp,"Unknown on-disk format, unable to mount\n" ); +if (sb->sb_bsize <512 || sb->sb_bsize >PAGE_SIZE || + (sb->sb_bsize & (sb->sb_bsize -1 ))) { +pr_warn ("Invalid superblock size\n" ); +return -EINVAL ; + } -return -EINVAL ; +return 0 ; }static void end_bio_io_page (struct bio *bio)
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 102/191 cifs: handle -EINTR in cifs_setattr
In reply to Greg Kroah-HartmanFrom: Ronnie Sahlberg <[email protected]>
[ Upstream commit c6cc4c5a72505a0ecefc9b413f16bec512f38078 ]
RHBZ: 1848178
Some calls that set attributes, like utimensat(), are not supposed to return -EINTR and thus do not have handlers for this in glibc which causes us to leak -EINTR to the applications which are also unprepared to handle it.
For example tar will break if utimensat() return -EINTR and abort unpacking the archive. Other applications may break too.
To handle this we add checks, and retry, for -EINTR in cifs_setattr()
--- fs/cifs/inode.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
fs/cifs/inode.c +9 -4
@@ -2550,13 +2550,18 @@ cifs_setattr(struct dentry *direntry, struct iattr *attrs) {struct cifs_sb_info *cifs_sb =CIFS_SB (direntry->d_sb);struct cifs_tcon *pTcon =cifs_sb_master_tcon (cifs_sb); +int rc, retries =0 ; -if (pTcon->unix_ext) -return cifs_setattr_unix (direntry, attrs); - -return cifs_setattr_nounix (direntry, attrs); +do { +if (pTcon->unix_ext) + rc =cifs_setattr_unix (direntry, attrs); +else + rc =cifs_setattr_nounix (direntry, attrs); + retries++; + }while (is_retryable_error (rc) && retries <2 );/* BB: add cifs_setattr_legacy for really old servers */ +return rc; }#if 0
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 103/191 arm64: dts: renesas: ulcb: add full-pwr-cycle-in-suspend into eMMC nodes
In reply to Greg Kroah-HartmanFrom: Yoshihiro Shimoda <[email protected]>
[ Upstream commit 992d7a8b88c83c05664b649fc54501ce58e19132 ]
Add full-pwr-cycle-in-suspend property to do a graceful shutdown of the eMMC device in system suspend.
--- arch/arm64/boot/dts/renesas/ulcb.dtsi | 1 + 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/renesas/ulcb.dtsi +1 -0
@@ -430,6 +430,7 @@ bus-width = <8>; mmc-hs200-1_8v; non-removable; + full-pwr-cycle-in-suspend; status = "okay"; };
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 104/191 ARM: dts: omap4: Fix sgx clock rate for 4430
In reply to Greg Kroah-HartmanFrom: Tony Lindgren <[email protected]>
[ Upstream commit 19d3e9a0bdd57b90175f30390edeb06851f5f9f3 ]
We currently have a different clock rate for droid4 compared to the stock v3.0.8 based Android Linux kernel:
# cat /sys/kernel/debug/clk/dpll_*_m7x2_ck/clk_rate 266666667 307200000 # cat /sys/kernel/debug/clk/l3_gfx_cm:clk:0000:0/clk_rate 307200000
Let's fix this by configuring sgx to use 153.6 MHz instead of 307.2 MHz. Looks like also at least duover needs this change to avoid hangs, so let's apply it for all 4430.
This helps a bit with thermal issues that seem to be related to memory corruption when using sgx. It seems that other driver related issues still remain though.
--- arch/arm/boot/dts/omap4.dtsi | 2 +- arch/arm/boot/dts/omap443x.dtsi | 10 ++++++++++ 2 files changed, 11 insertions(+), 1 deletion(-)
arch/arm/boot/dts/omap4.dtsi +1 -1
@@ -516,7 +516,7 @@ status = "disabled"; }; - target-module@56000000 { + sgx_module: target-module@56000000 { compatible = "ti,sysc-omap4", "ti,sysc"; ti,hwmods = "gpu"; reg = <0x5601fc00 0x4>,
arch/arm/boot/dts/omap443x.dtsi +10 -0
@@ -74,3 +74,13 @@ }; /include/ "omap443x-clocks.dtsi" + +/* + * Use dpll_per for sgx at 153.6MHz like droid4 stock v3.0.8 Android kernel + */ +&sgx_module { + assigned-clocks = <&l3_gfx_clkctrl OMAP4_GPU_CLKCTRL 24>, + <&dpll_per_m7x2_ck>; + assigned-clock-rates = <0>, <153600000>; + assigned-clock-parents = <&dpll_per_m7x2_ck>; +};
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 105/191 memory: emif: Remove bogus debugfs error handling
In reply to Greg Kroah-HartmanFrom: Dan Carpenter <[email protected]>
[ Upstream commit fd22781648080cc400772b3c68aa6b059d2d5420 ]
Callers are generally not supposed to check the return values from debugfs functions. Debugfs functions never return NULL so this error handling will never trigger. (Historically debugfs functions used to return a mix of NULL and error pointers but it was eventually deemed too complicated for something which wasn't intended to be used in normal situations).
Delete all the error handling.
--- drivers/memory/emif.c | 33 +++++---------------------------- 1 file changed, 5 insertions(+), 28 deletions(-)
drivers/memory/emif.c +5 -28
@@ -165,35 +165,12 @@ static const struct file_operations emif_mr4_fops = { static int __init_or_module emif_debugfs_init(struct emif_data *emif) { -struct dentry *dentry; -int ret; - - dentry =debugfs_create_dir (dev_name (emif->dev),NULL ); -if (!dentry) { - ret = -ENOMEM ; - goto err0; - } - emif->debugfs_root = dentry; - - dentry =debugfs_create_file ("regcache_dump" ,S_IRUGO , - emif->debugfs_root, emif, &emif_regdump_fops); -if (!dentry) { - ret = -ENOMEM ; - goto err1; - } - - dentry =debugfs_create_file ("mr4" ,S_IRUGO , - emif->debugfs_root, emif, &emif_mr4_fops); -if (!dentry) { - ret = -ENOMEM ; - goto err1; - } - + emif->debugfs_root =debugfs_create_dir (dev_name (emif->dev),NULL ); +debugfs_create_file ("regcache_dump" ,S_IRUGO , emif->debugfs_root, emif, + &emif_regdump_fops); +debugfs_create_file ("mr4" ,S_IRUGO , emif->debugfs_root, emif, + &emif_mr4_fops);return 0 ; -err1: -debugfs_remove_recursive (emif->debugfs_root); -err0: -return ret; } static void __exit emif_debugfs_exit(struct emif_data *emif)
Signature · 1 line
2.27.0
Re: [PATCH 4.19 105/191] memory: emif: Remove bogus debugfs error handling
Hi!
Quoted 4 lines
From: Dan Carpenter <[email protected]>
[ Upstream commit
fd22781648080cc400772b3c68aa6b059d2d5420]
Quoted 8 lines
Callers are generally not supposed to check the return values from debugfs functions. Debugfs functions never return NULL so this error handling will never trigger. (Historically debugfs functions used to return a mix of NULL and error pointers but it was eventually deemed too complicated for something which wasn't intended to be used in normal situations).
Delete all the error handling.
This is wrong for 4.19.
Memory functions still return NULL here.
Best regards,
Pavel
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Greg Kroah-Hartman
maintainerPATCH 106/191 ARM: dts: s5pv210: remove DMA controller bus node name to fix dtschema warnings
In reply to Greg Kroah-HartmanFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit ea4e792f3c8931fffec4d700cf6197d84e9f35a6 ]
There is no need to keep DMA controller nodes under AMBA bus node. Remove the "amba" node to fix dtschema warnings like:
amba: $nodename:0: 'amba' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 49 +++++++++++++++------------------- 1 file changed, 21 insertions(+), 28 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +21 -28
@@ -126,35 +126,28 @@ }; }; - amba { - #address-cells = <1>; - #size-cells = <1>; - compatible = "simple-bus"; - ranges; - - pdma0: dma@e0900000 { - compatible = "arm,pl330", "arm,primecell"; - reg = <0xe0900000 0x1000>; - interrupt-parent = <&vic0>; - interrupts = <19>; - clocks = <&clocks CLK_PDMA0>; - clock-names = "apb_pclk"; - #dma-cells = <1>; - #dma-channels = <8>; - #dma-requests = <32>; - }; + pdma0: dma@e0900000 { + compatible = "arm,pl330", "arm,primecell"; + reg = <0xe0900000 0x1000>; + interrupt-parent = <&vic0>; + interrupts = <19>; + clocks = <&clocks CLK_PDMA0>; + clock-names = "apb_pclk"; + #dma-cells = <1>; + #dma-channels = <8>; + #dma-requests = <32>; + }; - pdma1: dma@e0a00000 { - compatible = "arm,pl330", "arm,primecell"; - reg = <0xe0a00000 0x1000>; - interrupt-parent = <&vic0>; - interrupts = <20>; - clocks = <&clocks CLK_PDMA1>; - clock-names = "apb_pclk"; - #dma-cells = <1>; - #dma-channels = <8>; - #dma-requests = <32>; - }; + pdma1: dma@e0a00000 { + compatible = "arm,pl330", "arm,primecell"; + reg = <0xe0a00000 0x1000>; + interrupt-parent = <&vic0>; + interrupts = <20>; + clocks = <&clocks CLK_PDMA1>; + clock-names = "apb_pclk"; + #dma-cells = <1>; + #dma-channels = <8>; + #dma-requests = <32>; }; spi0: spi@e1300000 {
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 107/191 ARM: dts: s5pv210: move PMU node out of clock controller
In reply to Greg Kroah-HartmanFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit bb98fff84ad1ea321823759edaba573a16fa02bd ]
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and dtbs_check warnings like:
clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +5 -8
@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>; clock-names = "xxti", "xusbxti"; clocks = <&xxti>, <&xusbxti>; #clock-cells = <1>; - #address-cells = <1>; - #size-cells = <1>; - ranges; + }; - pmu_syscon: syscon@e0108000 { - compatible = "samsung-s5pv210-pmu", "syscon"; - reg = <0xe0108000 0x8000>; - }; + pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; }; pinctrl0: pinctrl@e0200000 {
Signature · 1 line
2.27.0
Re: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Hi!
Quoted 16 lines
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and
dtbs_checkwarnings like:clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'+++
b/arch/arm/boot/dts/s5pv210.dtsi(quoted diff) +1
-1@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>;
...
Review
Quoted 4 lines
(quoted diff) +3
-0+ pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; };
Should clock-controller@e0100000's reg be shortened to 0x8000 so that the ranges do not overlap?
Best regards,
Pavel
arch/arm/boot/dts/s5pv210.dtsi +1 -1
@@ -99,7 +99,7 @@ clocks: clock-controller@e0100000 { compatible = "samsung,s5pv210-clock"; - reg = <0xe0100000 0x10000>; + reg = <0xe0100000 0x8000>; clock-names = "xxti", "xusbxti"; clocks = <&xxti>, <&xusbxti>; #clock-cells = <1>;
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Krzysztof Kozlowski
maintainerRe: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Quoted 25 lines · Pavel Machek
Hi!
Quoted 16 lines
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and
dtbs_checkwarnings like:clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'+++
b/arch/arm/boot/dts/s5pv210.dtsi(quoted diff) +1
-1@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>;...
Review
Quoted 4 lines
(quoted diff) +3
-0+ pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; };
Quoted 4 lines
Should clock-controller@e0100000's reg be shortened to
0x8000so that the ranges do not overlap?Signed-off-by Pavel Machek (CIP) <[email protected]>
I don't think this commit should be backported to stable. It is simple dtbs_check - checking whether Devicetree source matches device tree schema. Neither the schema nor the warning existed in v4.19. I think dtbs_check fixes should not be backported, unless a real issue is pointed out.
Best regards,
Krzysztof
Re: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Hi!
Quoted 25 lines
Quoted 25 lines
Quoted 16 lines
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and
dtbs_checkwarnings like:clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'+++
b/arch/arm/boot/dts/s5pv210.dtsi(quoted diff) +1
-1@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>;...
Review
Quoted 4 lines
(quoted diff) +3
-0+ pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; };Should clock-controller@e0100000's reg be shortened to
0x8000so that the ranges do not overlap?
Quoted 7 lines
Quoted 1 line
Signed-off-by Pavel Machek (CIP) <[email protected]>I don't think this commit should be backported to stable. It is simple
dtbs_check- checking whether Devicetree source matches device tree schema. Neither the schema nor the warning existed in v4.19. I thinkdtbs_checkfixes should not be backported, unless a real issue is pointed out.
I agree with you about the backporting. Hopefully Greg drops the commit.
But the other issue is: should mainline be fixed so that ranges do not overlap?
Best regards,
Pavel
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Krzysztof Kozlowski
maintainerRe: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Quoted 35 lines · Pavel Machek
Hi!
Quoted 25 lines
Quoted 25 lines
Quoted 16 lines
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and
dtbs_checkwarnings like:clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'+++
b/arch/arm/boot/dts/s5pv210.dtsi(quoted diff) +1
-1@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>;...
Review
Quoted 4 lines
(quoted diff) +3
-0+ pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; };Should clock-controller@e0100000's reg be shortened to
0x8000so that the ranges do not overlap?Quoted 7 lines
Quoted 1 line
Signed-off-by Pavel Machek (CIP) <[email protected]>I don't think this commit should be backported to stable. It is simple
dtbs_check- checking whether Devicetree source matches device tree schema. Neither the schema nor the warning existed in v4.19. I thinkdtbs_checkfixes should not be backported, unless a real issue is pointed out.
Quoted 4 lines
I agree with you about the backporting. Hopefully Greg drops the commit.
But the other issue is: should mainline be fixed so that ranges do not overlap?
Yes, it should be. This should fail on mapping resources...
I'll take a look, thanks for the report.
Best regards,
Krzysztof
Krzysztof Kozlowski
maintainerRe: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Quoted 36 lines · Krzysztof Kozlowski
Quoted 35 lines · Pavel Machek
Hi!
Quoted 32 lines
Quoted 22 lines
Quoted 16 lines
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and
dtbs_checkwarnings like:clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'+++
b/arch/arm/boot/dts/s5pv210.dtsi(quoted diff) +1
-1@@ -98,19 +98,16 @@ }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>;...
Review
Quoted 4 lines
(quoted diff) +3
-0+ pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; };Quoted 4 lines
Should clock-controller@e0100000's reg be shortened to
0x8000so that the ranges do not overlap?Signed-off-by Pavel Machek (CIP) <[email protected]>I don't think this commit should be backported to stable. It is simple
dtbs_check- checking whether Devicetree source matches device tree schema. Neither the schema nor the warning existed in v4.19. I thinkdtbs_checkfixes should not be backported, unless a real issue is pointed out.
Quoted 8 lines
Quoted 4 lines
I agree with you about the backporting. Hopefully Greg drops the commit.
But the other issue is: should mainline be fixed so that ranges do not overlap?
Yes, it should be. This should fail on mapping resources...
I'll take a look, thanks for the report.
+Cc Paweł and Marek,
The IO memory mappings overlap unfortunately on purpose. Most of the clock driver registers are in the first range of 0x3000 but it also uses two registers at offset 0xe000.
The samsung-s5pv210-pmu is used only as a syscon by phy-s5pv210-usb2.c which wants to play with 0x680c.
The solution could be to split the mapping into two parts but I don't want to do this. I don't have the hardware so there is a chance I will break things.
However if Paweł, Jonathan or Marek want to improve it - patches are welcomed. :)
Best regards,
Krzysztof
Re: [PATCH 4.19 107/191] ARM: dts: s5pv210: move PMU node out of clock controller
Hi!
Quoted 24 lines
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
I don't think this commit should be backported to stable. It is simple
dtbs_check- checking whether Devicetree source matches device tree schema. Neither the schema nor the warning existed in v4.19. I thinkdtbs_checkfixes should not be backported, unless a real issue is pointed out.Quoted 8 lines
Quoted 4 lines
I agree with you about the backporting. Hopefully Greg drops the commit.
But the other issue is: should mainline be fixed so that ranges do not overlap?
Yes, it should be. This should fail on mapping resources...
I'll take a look, thanks for the report.
+Cc Paweł and Marek,
The IO memory mappings overlap unfortunately on purpose. Most of the clock driver registers are in the first range of
0x3000but it also uses two registers at offset0xe000.The samsung-s5pv210-pmu is used only as a syscon by
phy-s5pv210-usb2.cwhich wants to play with0x680c.
Quoted 6 lines
The solution could be to split the mapping into two parts but I don't want to do this. I don't have the hardware so there is a chance I will break things.
However if Paweł, Jonathan or Marek want to improve it - patches are welcomed. :)
Okay, it would be nice to at least have a comment there.
Best regards,
Pavel
Signature · 1 line
http://www.livejournal.com/~pavelmachek
Greg Kroah-Hartman
maintainerPATCH 108/191 ARM: dts: s5pv210: remove dedicated audio-subsystem node
In reply to Greg Kroah-HartmanFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 6c17a2974abf68a58517f75741b15c4aba42b4b8 ]
The 'audio-subsystem' node is an artificial creation, not representing real hardware. The hardware is described by its nodes - AUDSS clock controller and I2S0.
Remove the 'audio-subsystem' node along with its undocumented compatible to fix dtbs_check warnings like:
audio-subsystem: $nodename:0: 'audio-subsystem' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 65 +++++++++++++++------------------- 1 file changed, 29 insertions(+), 36 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +29 -36
@@ -217,43 +217,36 @@ status = "disabled"; }; - audio-subsystem { - compatible = "samsung,s5pv210-audss", "simple-bus"; - #address-cells = <1>; - #size-cells = <1>; - ranges; - - clk_audss: clock-controller@eee10000 { - compatible = "samsung,s5pv210-audss-clock"; - reg = <0xeee10000 0x1000>; - clock-names = "hclk", "xxti", - "fout_epll", - "sclk_audio0"; - clocks = <&clocks DOUT_HCLKP>, <&xxti>, - <&clocks FOUT_EPLL>, - <&clocks SCLK_AUDIO0>; - #clock-cells = <1>; - }; + clk_audss: clock-controller@eee10000 { + compatible = "samsung,s5pv210-audss-clock"; + reg = <0xeee10000 0x1000>; + clock-names = "hclk", "xxti", + "fout_epll", + "sclk_audio0"; + clocks = <&clocks DOUT_HCLKP>, <&xxti>, + <&clocks FOUT_EPLL>, + <&clocks SCLK_AUDIO0>; + #clock-cells = <1>; + }; - i2s0: i2s@eee30000 { - compatible = "samsung,s5pv210-i2s"; - reg = <0xeee30000 0x1000>; - interrupt-parent = <&vic2>; - interrupts = <16>; - dma-names = "rx", "tx", "tx-sec"; - dmas = <&pdma1 9>, <&pdma1 10>, <&pdma1 11>; - clock-names = "iis", - "i2s_opclk0", - "i2s_opclk1"; - clocks = <&clk_audss CLK_I2S>, - <&clk_audss CLK_I2S>, - <&clk_audss CLK_DOUT_AUD_BUS>; - samsung,idma-addr = <0xc0010000>; - pinctrl-names = "default"; - pinctrl-0 = <&i2s0_bus>; - #sound-dai-cells = <0>; - status = "disabled"; - }; + i2s0: i2s@eee30000 { + compatible = "samsung,s5pv210-i2s"; + reg = <0xeee30000 0x1000>; + interrupt-parent = <&vic2>; + interrupts = <16>; + dma-names = "rx", "tx", "tx-sec"; + dmas = <&pdma1 9>, <&pdma1 10>, <&pdma1 11>; + clock-names = "iis", + "i2s_opclk0", + "i2s_opclk1"; + clocks = <&clk_audss CLK_I2S>, + <&clk_audss CLK_I2S>, + <&clk_audss CLK_DOUT_AUD_BUS>; + samsung,idma-addr = <0xc0010000>; + pinctrl-names = "default"; + pinctrl-0 = <&i2s0_bus>; + #sound-dai-cells = <0>; + status = "disabled"; }; i2s1: i2s@e2100000 {
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 109/191 nbd: make the config put is called before the notifying the waiter
In reply to Greg Kroah-HartmanFrom: Xiubo Li <[email protected]>
[ Upstream commit 87aac3a80af5cbad93e63250e8a1e19095ba0d30 ]
There has one race case for ceph's rbd-nbd tool. When do mapping it may fail with EBUSY from ioctl(nbd, NBD_DO_IT), but actually the nbd device has already unmaped.
It dues to if just after the wake_up(), the recv_work() is scheduled out and defers calling the nbd_config_put(), though the map process has exited the "nbd->recv_task" is not cleared.
--- drivers/block/nbd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/block/nbd.c +1 -1
@@ -740,9 +740,9 @@ static void recv_work(struct work_struct *work)blk_mq_complete_request (blk_mq_rq_from_pdu (cmd)); } +nbd_config_put (nbd);atomic_dec (&config->recv_threads);wake_up (&config->recv_wq); -nbd_config_put (nbd);kfree (args); }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 110/191 sgl_alloc_order: fix memory leak
In reply to Greg Kroah-HartmanFrom: Douglas Gilbert <[email protected]>
[ Upstream commit b2a182a40278bc5849730e66bca01a762188ed86 ]
sgl_alloc_order() can fail when 'length' is large on a memory constrained system. When order > 0 it will potentially be making several multi-page allocations with the later ones more likely to fail than the earlier one. So it is important that sgl_alloc_order() frees up any pages it has obtained before returning NULL. In the case when order > 0 it calls the wrong free page function and leaks. In testing the leak was sufficient to bring down my 8 GiB laptop with OOM.
--- lib/scatterlist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
lib/scatterlist.c +1 -1
@@ -506,7 +506,7 @@ struct scatterlist *sgl_alloc_order(unsigned long long length, elem_len =min_t (u64, length,PAGE_SIZE << order); page =alloc_pages (gfp, order);if (!page) { -sgl_free (sgl); +sgl_free_order (sgl, order);return NULL ; }
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 111/191 nvme-rdma: fix crash when connect rejected
In reply to Greg Kroah-HartmanFrom: Chao Leng <[email protected]>
[ Upstream commit 43efdb8e870ee0f58633fd579aa5b5185bf5d39e ]
A crash can happened when a connect is rejected. The host establishes
the connection after received ConnectReply, and then continues to send the fabrics Connect command. If the controller does not receive the ReadyToUse capsule, host may receive a ConnectReject reply.
Call nvme_rdma_destroy_queue_ib after the host received the RDMA_CM_EVENT_REJECTED event. Then when the fabrics Connect command times out, nvme_rdma_timeout calls nvme_rdma_complete_rq to fail the request. A crash happenes due to use after free in nvme_rdma_complete_rq.
nvme_rdma_destroy_queue_ib is redundant when handling the RDMA_CM_EVENT_REJECTED event as nvme_rdma_destroy_queue_ib is already called in connection failure handler.
--- drivers/nvme/host/rdma.c | 1 - 1 file changed, 1 deletion(-)
drivers/nvme/host/rdma.c +0 -1
@@ -1640,7 +1640,6 @@ static int nvme_rdma_cm_handler(struct rdma_cm_id *cm_id,complete (&queue->cm_done);return 0 ;case RDMA_CM_EVENT_REJECTED : -nvme_rdma_destroy_queue_ib (queue); cm_error =nvme_rdma_conn_rejected (queue, ev);break ;case RDMA_CM_EVENT_ROUTE_ERROR :
Signature · 1 line
2.27.0
Greg Kroah-Hartman
maintainerPATCH 112/191 md/raid5: fix oops during stripe resizing
In reply to Greg Kroah-HartmanFrom: Song Liu <[email protected]>
commit b44c018cdf748b96b676ba09fdbc5b34fc443ada upstream.
KoWei reported crash during raid5 reshape:
[ 1032.252932] Oops: 0002 [#1] SMP PTI [...] [ 1032.252943] RIP: 0010:memcpy_erms+0x6/0x10 [...] [ 1032.252947] RSP: 0018:ffffba1ac0c03b78 EFLAGS: 00010286 [ 1032.252949] RAX: 0000784ac0000000 RBX: ffff91bec3d09740 RCX: 0000000000001000 [ 1032.252951] RDX: 0000000000001000 RSI: ffff91be6781c000 RDI: 0000784ac0000000 [ 1032.252953] RBP: ffffba1ac0c03bd8 R08: 0000000000001000 R09: ffffba1ac0c03bf8 [ 1032.252954] R10: 0000000000000000 R11: 0000000000000000 R12: ffffba1ac0c03bf8 [ 1032.252955] R13: 0000000000001000 R14: 0000000000000000 R15: 0000000000000000 [ 1032.252958] FS: 0000000000000000(0000) GS:ffff91becf500000(0000) knlGS:0000000000000000 [ 1032.252959] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1032.252961] CR2: 0000784ac0000000 CR3: 000000031780a002 CR4: 00000000001606e0 [ 1032.252962] Call Trace: [ 1032.252969] ? async_memcpy+0x179/0x1000 [async_memcpy] [ 1032.252977] ? raid5_release_stripe+0x8e/0x110 [raid456] [ 1032.252982] handle_stripe_expansion+0x15a/0x1f0 [raid456] [ 1032.252988] handle_stripe+0x592/0x1270 [raid456] [ 1032.252993] handle_active_stripes.isra.0+0x3cb/0x5a0 [raid456] [ 1032.252999] raid5d+0x35c/0x550 [raid456] [ 1032.253002] ? schedule+0x42/0xb0 [ 1032.253006] ? schedule_timeout+0x10e/0x160 [ 1032.253011] md_thread+0x97/0x160 [ 1032.253015] ? wait_woken+0x80/0x80 [ 1032.253019] kthread+0x104/0x140 [ 1032.253022] ? md_start_sync+0x60/0x60 [ 1032.253024] ? kthread_park+0x90/0x90 [ 1032.253027] ret_from_fork+0x35/0x40
This is because cache_size_mutex was unlocked too early in resize_stripes, which races with grow_one_stripe() that grow_one_stripe() allocates a stripe with wrong pool_size.
Fix this issue by unlocking cache_size_mutex after updating pool_size.
--- drivers/md/raid5.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/raid5.c +2 -2
@@ -2417,8 +2417,6 @@ static int resize_stripes(struct r5conf }else err = -ENOMEM ; -mutex_unlock (&conf->cache_size_mutex); - conf->slab_cache = sc; conf->active_name =1 -conf->active_name;
@@ -2441,6 +2439,8 @@ static int resize_stripes(struct r5confif (!err) conf->pool_size = newsize; +mutex_unlock (&conf->cache_size_mutex); +return err; }
Greg Kroah-Hartman
maintainerPATCH 113/191 mmc: sdhci-acpi: AMDI0040: Set SDHCI_QUIRK2_PRESET_VALUE_BROKEN
In reply to Greg Kroah-HartmanFrom: Raul E Rangel <[email protected]>
commit f23cc3ba491af77395cea3f9d51204398729f26b upstream.
This change fixes HS400 tuning for devices with invalid presets.
SDHCI presets are not currently used for eMMC HS/HS200/HS400, but are used for DDR52. The HS400 retuning sequence is:
HS400->DDR52->HS->HS200->Perform Tuning->HS->HS400
This means that when HS400 tuning happens, we transition through DDR52 for a very brief period. This causes presets to be enabled unintentionally and stay enabled when transitioning back to HS200 or HS400. Some firmware has invalid presets, so we end up with driver strengths that can cause I/O problems.
--- drivers/mmc/host/sdhci-acpi.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+)
drivers/mmc/host/sdhci-acpi.c +37 -0
@@ -546,6 +546,43 @@ static int sdhci_acpi_emmc_amd_probe_slo (host->mmc->caps & MMC_CAP_1_8V_DDR)) host->mmc->caps2 =MMC_CAP2_HS400_1_8V ; +/* +* There are two types of presets out in the wild: +* 1) Default/broken presets. +* These presets have two sets of problems: +* a) The clock divisor for SDR12, SDR25, and SDR50 is too small. +* This results in clock frequencies that are 2x higher than +* acceptable. i.e., SDR12 = 25 MHz, SDR25 = 50 MHz, SDR50 = +* 100 MHz.x +* b) The HS200 and HS400 driver strengths don't match. +* By default, the SDR104 preset register has a driver strength of +* A, but the (internal) HS400 preset register has a driver +* strength of B. As part of initializing HS400, HS200 tuning +* needs to be performed. Having different driver strengths +* between tuning and operation is wrong. It results in different +* rise/fall times that lead to incorrect sampling. +* 2) Firmware with properly initialized presets. +* These presets have proper clock divisors. i.e., SDR12 => 12MHz, +* SDR25 => 25 MHz, SDR50 => 50 MHz. Additionally the HS200 and +* HS400 preset driver strengths match. +* +* Enabling presets for HS400 doesn't work for the following reasons: +* 1) sdhci_set_ios has a hard coded list of timings that are used +* to determine if presets should be enabled. +* 2) sdhci_get_preset_value is using a non-standard register to +* read out HS400 presets. The AMD controller doesn't support this +* non-standard register. In fact, it doesn't expose the HS400 +* preset register anywhere in the SDHCI memory map. This results +* in reading a garbage value and using the wrong presets. +* +* Since HS400 and HS200 presets must be identical, we could +* instead use the the SDR104 preset register. +* +* If the above issues are resolved we could remove this quirk for +* firmware that that has valid presets (i.e., SDR12 <= 12 MHz). +*/ + host->quirks2 |=SDHCI_QUIRK2_PRESET_VALUE_BROKEN ; + host->mmc_host_ops.select_drive_strength = amd_select_drive_strength; host->mmc_host_ops.set_ios = amd_set_ios;return 0 ;
Greg Kroah-Hartman
maintainerPATCH 114/191 perf/x86/amd/ibs: Dont include randomized bits in get_ibs_op_count()
In reply to Greg Kroah-HartmanFrom: Kim Phillips <[email protected]>
commit 680d69635005ba0e58fe3f4c52fc162b8fc743b0 upstream.
get_ibs_op_count() adds hardware's current count (IbsOpCurCnt) bits to its count regardless of hardware's valid status.
According to the PPR for AMD Family 17h Model 31h B0 55803 Rev 0.54, if the counter rolls over, valid status is set, and the lower 7 bits of IbsOpCurCnt are randomized by hardware.
Don't include those bits in the driver's event count.
--- arch/x86/events/amd/ibs.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)
arch/x86/events/amd/ibs.c +8 -4
@@ -347,11 +347,15 @@ static u64 get_ibs_op_count(u64 config) {u64 count =0 ; +/* +* If the internal 27-bit counter rolled over, the count is MaxCnt +* and the lower 7 bits of CurCnt are randomized. +* Otherwise CurCnt has the full 27-bit current counter value. +*/ if (config &IBS_OP_VAL ) - count += (config &IBS_OP_MAX_CNT ) <<4 ;/* cnt rolled over */ - -if (ibs_caps &IBS_CAPS_RDWROPCNT ) - count += (config &IBS_OP_CUR_CNT ) >>32 ; + count = (config &IBS_OP_MAX_CNT ) <<4 ; +else if (ibs_caps &IBS_CAPS_RDWROPCNT ) + count = (config &IBS_OP_CUR_CNT ) >>32 ;return count; }
Greg Kroah-Hartman
maintainerPATCH 115/191 perf/x86/amd/ibs: Fix raw sample data accumulation
In reply to Greg Kroah-HartmanFrom: Kim Phillips <[email protected]>
commit 36e1be8ada994d509538b3b1d0af8b63c351e729 upstream.
Neither IbsBrTarget nor OPDATA4 are populated in IBS Fetch mode. Don't accumulate them into raw sample user data in that case.
Also, in Fetch mode, add saving the IBS Fetch Control Extended MSR.
Technically, there is an ABI change here with respect to the IBS raw sample data format, but I don't see any perf driver version information being included in perf.data file headers, but, existing users can detect whether the size of the sample record has reduced by 8 bytes to determine whether the IBS driver has this fix.
--- arch/x86/events/amd/ibs.c | 26 ++++++++++++++++---------- arch/x86/include/asm/msr-index.h | 1 + 2 files changed, 17 insertions(+), 10 deletions(-)
arch/x86/events/amd/ibs.c +16 -10
@@ -647,18 +647,24 @@ fail: perf_ibs->offset_max, offset + 1); }while (offset < offset_max); +/* +* Read IbsBrTarget, IbsOpData4, and IbsExtdCtl separately +* depending on their availability. +* Can't add to offset_max as they are staggered +*/ if (event->attr.sample_type &PERF_SAMPLE_RAW ) { -/* -* Read IbsBrTarget and IbsOpData4 separately -* depending on their availability. -* Can't add to offset_max as they are staggered -*/ -if (ibs_caps &IBS_CAPS_BRNTRGT ) { -rdmsrl (MSR_AMD64_IBSBRTARGET , *buf++); - size++; +if (perf_ibs == &perf_ibs_op) { +if (ibs_caps &IBS_CAPS_BRNTRGT ) { +rdmsrl (MSR_AMD64_IBSBRTARGET , *buf++); + size++; + } +if (ibs_caps &IBS_CAPS_OPDATA4 ) { +rdmsrl (MSR_AMD64_IBSOPDATA4 , *buf++); + size++; + } } -if (ibs_caps &IBS_CAPS_OPDATA4 ) { -rdmsrl (MSR_AMD64_IBSOPDATA4 , *buf++); +if (perf_ibs == &perf_ibs_fetch && (ibs_caps &IBS_CAPS_FETCHCTLEXTD )) { +rdmsrl (MSR_AMD64_ICIBSEXTDCTL , *buf++); size++; } }
arch/x86/include/asm/msr-index.h +1 -0
@@ -377,6 +377,7 @@#define MSR_AMD64_IBSOP_REG_MASK ((1UL<<MSR_AMD64_IBSOP_REG_COUNT)-1)#define MSR_AMD64_IBSCTL 0xc001103a#define MSR_AMD64_IBSBRTARGET 0xc001103b +#define MSR_AMD64_ICIBSEXTDCTL 0xc001103c#define MSR_AMD64_IBSOPDATA4 0xc001103d#define MSR_AMD64_IBS_REG_COUNT_MAX 8/* includes MSR_AMD64_IBSBRTARGET */ #define MSR_AMD64_SEV 0xc0010131
Greg Kroah-Hartman
maintainerPATCH 116/191 leds: bcm6328, bcm6358: use devres LED registering function
In reply to Greg Kroah-HartmanFrom: Marek Behún <[email protected]>
commit ff5c89d44453e7ad99502b04bf798a3fc32c758b upstream.
These two drivers do not provide remove method and use devres for allocation of other resources, yet they use led_classdev_register instead of the devres variant, devm_led_classdev_register.
Fix this.
--- drivers/leds/leds-bcm6328.c | 2 +- drivers/leds/leds-bcm6358.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/leds/leds-bcm6328.c +1 -1
@@ -336,7 +336,7 @@ static int bcm6328_led(struct device *de led->cdev.brightness_set = bcm6328_led_set; led->cdev.blink_set = bcm6328_blink_set; - rc =led_classdev_register (dev, &led->cdev); + rc =devm_led_classdev_register (dev, &led->cdev);if (rc <0 )return rc;
drivers/leds/leds-bcm6358.c +1 -1
@@ -141,7 +141,7 @@ static int bcm6358_led(struct device *de led->cdev.brightness_set = bcm6358_led_set; - rc =led_classdev_register (dev, &led->cdev); + rc =devm_led_classdev_register (dev, &led->cdev);if (rc <0 )return rc;
Greg Kroah-Hartman
maintainerPATCH 117/191 media: uvcvideo: Fix uvc_ctrl_fixup_xu_info() not having any effect
In reply to Greg Kroah-HartmanFrom: Hans de Goede <[email protected]>
commit 93df48d37c3f03886d84831992926333e7810640 upstream.
uvc_ctrl_add_info() calls uvc_ctrl_get_flags() which will override the fixed-up flags set by uvc_ctrl_fixup_xu_info().
uvc_ctrl_init_xu_ctrl() already calls uvc_ctrl_get_flags() before calling uvc_ctrl_add_info(), so the uvc_ctrl_get_flags() call in uvc_ctrl_add_info() is not necessary for xu ctrls.
This commit moves the uvc_ctrl_get_flags() call for normal controls from uvc_ctrl_add_info() to uvc_ctrl_init_ctrl(), so that we no longer call uvc_ctrl_get_flags() twice for xu controls and so that we no longer override the fixed-up flags set by uvc_ctrl_fixup_xu_info().
This fixes the xu motor controls not working properly on a Logitech 046d:08cc, and presumably also on the other Logitech models which have a quirk for this in the uvc_ctrl_fixup_xu_info() function.
--- drivers/media/usb/uvc/uvc_ctrl.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/media/usb/uvc/uvc_ctrl.c +7 -7
@@ -2038,13 +2038,6 @@ static int uvc_ctrl_add_info(struct uvc_ goto done; } -/* -* Retrieve control flags from the device. Ignore errors and work with -* default flag values from the uvc_ctrl array when the device doesn't -* properly implement GET_INFO on standard controls. -*/ -uvc_ctrl_get_flags (dev, ctrl, &ctrl->info); - ctrl->initialized =1 ; uvc_trace(UVC_TRACE_CONTROL, "Added control %pUl/%u to device %s "
@@ -2267,6 +2260,13 @@ static void uvc_ctrl_init_ctrl(struct uvif (uvc_entity_match_guid (ctrl->entity, info->entity) && ctrl->index == info->index) {uvc_ctrl_add_info (dev, ctrl, info); +/* +* Retrieve control flags from the device. Ignore errors +* and work with default flag values from the uvc_ctrl +* array when the device doesn't properly implement +* GET_INFO on standard controls. +*/ +uvc_ctrl_get_flags (dev, ctrl, &ctrl->info);break ; } }
Greg Kroah-Hartman
maintainerPATCH 118/191 fs: Dont invalidate page buffers in block_write_full_page()
In reply to Greg Kroah-HartmanFrom: Jan Kara <[email protected]>
commit 6dbf7bb555981fb5faf7b691e8f6169fc2b2e63b upstream.
If block_write_full_page() is called for a page that is beyond current inode size, it will truncate page buffers for the page and return 0. This logic has been added in 2.5.62 in commit 81eb69062588 ("fix ext3 BUG due to race with truncate") in history.git tree to fix a problem with ext3 in data=ordered mode. This particular problem doesn't exist anymore because ext3 is long gone and ext4 handles ordered data differently. Also normally buffers are invalidated by truncate code and there's no need to specially handle this in ->writepage() code.
This invalidation of page buffers in block_write_full_page() is causing issues to filesystems (e.g. ext4 or ocfs2) when block device is shrunk under filesystem's hands and metadata buffers get discarded while being tracked by the journalling layer. Although it is obviously "not supported" it can cause kernel crashes like:
[ 7986.689400] BUG: unable to handle kernel NULL pointer dereference at +0000000000000008 [ 7986.697197] PGD 0 P4D 0 [ 7986.699724] Oops: 0002 [#1] SMP PTI [ 7986.703200] CPU: 4 PID: 203778 Comm: jbd2/dm-3-8 Kdump: loaded Tainted: G +O --------- - - 4.18.0-147.5.0.5.h126.eulerosv2r9.x86_64 #1 [ 7986.716438] Hardware name: Huawei RH2288H V3/BC11HGSA0, BIOS 1.57 08/11/2015 [ 7986.723462] RIP: 0010:jbd2_journal_grab_journal_head+0x1b/0x40 [jbd2] ... [ 7986.810150] Call Trace: [ 7986.812595] __jbd2_journal_insert_checkpoint+0x23/0x70 [jbd2] [ 7986.818408] jbd2_journal_commit_transaction+0x155f/0x1b60 [jbd2] [ 7986.836467] kjournald2+0xbd/0x270 [jbd2]
which is not great. The crash happens because bh->b_private is suddently NULL although BH_JBD flag is still set (this is because block_invalidatepage() cleared BH_Mapped flag and subsequent bh lookup found buffer without BH_Mapped set, called init_page_buffers() which has rewritten bh->b_private). So just remove the invalidation in block_write_full_page().
Note that the buffer cache invalidation when block device changes size is already careful to avoid similar problems by using invalidate_mapping_pages() which skips busy buffers so it was only this odd block_write_full_page() behavior that could tear down bdev buffers under filesystem's hands.
--- fs/buffer.c | 16 ---------------- 1 file changed, 16 deletions(-)
fs/buffer.c +0 -16
@@ -2742,16 +2742,6 @@ int nobh_writepage(struct page *page, ge/* Is the page fully outside i_size? (truncate in progress) */ offset = i_size & (PAGE_SIZE -1 );if (page->index >= end_index+1 || !offset) { -/* -* The page may have dirty, unmapped buffers. For example, -* they may have been added in ext3_writepage(). Make them -* freeable here, so the page does not leak. -*/ -#if 0 -/* Not really sure about this - do we need this ? */ -if (page->mapping->a_ops->invalidatepage) - page->mapping->a_ops->invalidatepage (page, offset); -#endif unlock_page (page);return 0 ;/* don't care */ }
@@ -2946,12 +2936,6 @@ int block_write_full_page(struct page *p/* Is the page fully outside i_size? (truncate in progress) */ offset = i_size & (PAGE_SIZE -1 );if (page->index >= end_index+1 || !offset) { -/* -* The page may have dirty, unmapped buffers. For example, -* they may have been added in ext3_writepage(). Make them -* freeable here, so the page does not leak. -*/ -do_invalidatepage (page,0 ,PAGE_SIZE );unlock_page (page);return 0 ;/* don't care */ }
Greg Kroah-Hartman
maintainerPATCH 119/191 NFS: fix nfs_path in case of a rename retry
In reply to Greg Kroah-HartmanFrom: Ashish Sangwan <[email protected]>
commit 247db73560bc3e5aef6db50c443c3c0db115bc93 upstream.
We are generating incorrect path in case of rename retry because we are restarting from wrong dentry. We should restart from the dentry which was received in the call to nfs_path.
--- fs/nfs/namespace.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)
fs/nfs/namespace.c +8 -4
@@ -30,9 +30,9 @@ int nfs_mountpoint_expiry_timeout = 500/* * nfs_path - reconstruct the path given an arbitrary dentry * @base - used to return pointer to the end of devname part of path -* @dentry - pointer to dentry +* @dentry_in - pointer to dentry * @buffer - result buffer -* @buflen - length of buffer +* @buflen_in - length of buffer * @flags - options (see below) * * Helper function for constructing the server pathname
@@ -47,15 +47,19 @@ int nfs_mountpoint_expiry_timeout = 500* the original device (export) name * (if unset, the original name is returned verbatim) */ -char *nfs_path (char **p,struct dentry *dentry,char *buffer,ssize_t buflen, -unsigned flags) +char *nfs_path (char **p,struct dentry *dentry_in,char *buffer, +ssize_t buflen_in,unsigned flags) {char *end;int namelen;unsigned seq;const char *base; +struct dentry *dentry; +ssize_t buflen; rename_retry: + buflen = buflen_in; + dentry = dentry_in; end = buffer+buflen; *--end ='\0' ; buflen--;
Greg Kroah-Hartman
maintainerPATCH 120/191 ACPI: button: fix handling lid state changes when input device closed
In reply to Greg Kroah-HartmanFrom: [email protected] <[email protected]>
commit 21988a8e51479ceffe7b0568b170effabb708dfe upstream.
The original intent of 84d3f6b76447 was to delay evaluating lid state until all drivers have been loaded, with input device being opened from userspace serving as a signal for this condition. Let's ensure that state updates happen even if userspace closed (or in the future inhibited) input device.
Note that if we go through suspend/resume cycle we assume the system has been fully initialized even if LID input device has not been opened yet.
This has a side-effect of fixing access to input->users outside of input->mutex protections by the way of eliminating said accesses and using driver private flag.
--- drivers/acpi/button.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/acpi/button.c +7 -6
@@ -149,6 +149,7 @@ struct acpi_button {int last_state;ktime_t last_time;bool suspended; +bool lid_state_initialized; };static BLOCKING_NOTIFIER_HEAD (acpi_lid_notifier);
@@ -404,6 +405,8 @@ static int acpi_lid_update_state(structstatic void acpi_lid_initialize_state (struct acpi_device *device) { +struct acpi_button *button =acpi_driver_data (device); +switch (lid_init_state) {case ACPI_BUTTON_LID_INIT_OPEN : (void )acpi_lid_notify_state (device,1 );
@@ -415,13 +418,14 @@ static void acpi_lid_initialize_state(stdefault :break ; } + + button->lid_state_initialized = true; }static void acpi_button_notify (struct acpi_device *device,u32 event) {struct acpi_button *button =acpi_driver_data (device);struct input_dev *input; -int users;switch (event) {case ACPI_FIXED_HARDWARE_EVENT :
@@ -430,10 +434,7 @@ static void acpi_button_notify(struct accase ACPI_BUTTON_NOTIFY_STATUS : input = button->input;if (button->type ==ACPI_BUTTON_TYPE_LID ) { -mutex_lock (&button->input->mutex); - users = button->input->users; -mutex_unlock (&button->input->mutex); -if (users) +if (button->lid_state_initialized)acpi_lid_update_state (device, true); }else {int keycode;
@@ -478,7 +479,7 @@ static int acpi_button_resume(struct devstruct acpi_button *button =acpi_driver_data (device); button->suspended = false; -if (button->type ==ACPI_BUTTON_TYPE_LID && button->input->users) { +if (button->type ==ACPI_BUTTON_TYPE_LID ) { button->last_state = !!acpi_lid_evaluate_state (device); button->last_time =ktime_get ();acpi_lid_initialize_state (device);
Greg Kroah-Hartman
maintainerPATCH 121/191 ACPI / extlog: Check for RDMSR failure
In reply to Greg Kroah-HartmanFrom: Ben Hutchings <[email protected]>
commit 7cecb47f55e00282f972a1e0b09136c8cd938221 upstream.
extlog_init() uses rdmsrl() to read an MSR, which on older CPUs provokes a error message at boot:
unchecked MSR access error: RDMSR from 0x179 at rIP: 0xcd047307 (native_read_msr+0x7/0x40)
Use rdmsrl_safe() instead, and return -ENODEV if it fails.
--- drivers/acpi/acpi_extlog.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/acpi/acpi_extlog.c +3 -3
@@ -224,9 +224,9 @@ static int __init extlog_init(void)u64 cap;int rc; -rdmsrl (MSR_IA32_MCG_CAP , cap); - -if (!(cap &MCG_ELOG_P ) || !extlog_get_l1addr ()) +if (rdmsrl_safe (MSR_IA32_MCG_CAP , &cap) || + !(cap &MCG_ELOG_P ) || + !extlog_get_l1addr ())return -ENODEV ;if (edac_get_report_status () ==EDAC_REPORTING_FORCE ) {
Greg Kroah-Hartman
maintainerPATCH 122/191 ACPI: video: use ACPI backlight for HP 635 Notebook
In reply to Greg Kroah-HartmanFrom: Alex Hung <[email protected]>
commit b226faab4e7890bbbccdf794e8b94276414f9058 upstream.
The default backlight interface is AMD's radeon_bl0 which does not work on this system, so use the ACPI backlight interface on it instead.
--- drivers/acpi/video_detect.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/acpi/video_detect.c +9 -0
@@ -274,6 +274,15 @@ static const struct dmi_system_id video_ DMI_MATCH(DMI_PRODUCT_NAME, "530U4E/540U4E"), }, }, +/* https://bugs.launchpad.net/bugs/1894667 */ + { + .callback = video_detect_force_video, + .ident = "HP 635 Notebook", + .matches = { +DMI_MATCH (DMI_SYS_VENDOR ,"Hewlett-Packard" ), + DMI_MATCH(DMI_PRODUCT_NAME, "HP 635 Notebook PC"), + }, + }, /* Non win8 machines which need native backlight nevertheless */ {
Greg Kroah-Hartman
maintainerPATCH 123/191 ACPI: debug: dont allow debugging when ACPI is disabled
In reply to Greg Kroah-HartmanFrom: Jamie Iles <[email protected]>
commit 0fada277147ffc6d694aa32162f51198d4f10d94 upstream.
If ACPI is disabled then loading the acpi_dbg module will result in the following splat when lock debugging is enabled.
DEBUG_LOCKS_WARN_ON(lock->magic != lock)
WARNING: CPU: 0 PID: 1 at kernel/locking/mutex.c:938 __mutex_lock+0xa10/0x1290 Kernel panic - not syncing: panic_on_warn set ... CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc8+ #103 Hardware name: linux,dummy-virt (DT) Call trace: dump_backtrace+0x0/0x4d8 show_stack+0x34/0x48 dump_stack+0x174/0x1f8 panic+0x360/0x7a0 __warn+0x244/0x2ec report_bug+0x240/0x398 bug_handler+0x50/0xc0 call_break_hook+0x160/0x1d8 brk_handler+0x30/0xc0 do_debug_exception+0x184/0x340 el1_dbg+0x48/0xb0 el1_sync_handler+0x170/0x1c8 el1_sync+0x80/0x100 __mutex_lock+0xa10/0x1290 mutex_lock_nested+0x6c/0xc0 acpi_register_debugger+0x40/0x88 acpi_aml_init+0xc4/0x114 do_one_initcall+0x24c/0xb10 kernel_init_freeable+0x690/0x728 kernel_init+0x20/0x1e8 ret_from_fork+0x10/0x18
This is because acpi_debugger.lock has not been initialized as acpi_debugger_init() is not called when ACPI is disabled. Fail module loading to avoid this and any subsequent problems that might arise by trying to debug AML when ACPI is disabled.
--- drivers/acpi/acpi_dbg.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/acpi/acpi_dbg.c +3 -0
@@ -757,6 +757,9 @@ int __init acpi_aml_init(void) goto err_exit; } +if (acpi_disabled) +return -ENODEV ; +/* Initialize AML IO interface */ mutex_init (&acpi_aml_io.lock);init_waitqueue_head (&acpi_aml_io.wait);
Greg Kroah-Hartman
maintainerPATCH 124/191 acpi-cpufreq: Honor _PSD table setting on new AMD CPUs
In reply to Greg Kroah-HartmanFrom: Wei Huang <[email protected]>
commit 5368512abe08a28525d9b24abbfc2a72493e8dba upstream.
acpi-cpufreq has a old quirk that overrides the _PSD table supplied by BIOS on AMD CPUs. However the _PSD table of new AMD CPUs (Family 19h+) now accurately reports the P-state dependency of CPU cores. Hence this quirk needs to be fixed in order to support new CPUs' frequency control.
--- drivers/cpufreq/acpi-cpufreq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/cpufreq/acpi-cpufreq.c +2 -1
@@ -701,7 +701,8 @@ static int acpi_cpufreq_cpu_init(structcpumask_copy (policy->cpus,topology_core_cpumask (cpu)); } -if (check_amd_hwpstate_cpu (cpu) && !acpi_pstate_strict) { +if (check_amd_hwpstate_cpu (cpu) && boot_cpu_data.x86 <0x19 && + !acpi_pstate_strict) {cpumask_clear (policy->cpus);cpumask_set_cpu (cpu, policy->cpus); cpumask_copy(data->freqdomain_cpus,
Greg Kroah-Hartman
maintainerPATCH 125/191 w1: mxc_w1: Fix timeout resolution problem leading to bus error
In reply to Greg Kroah-HartmanFrom: Martin Fuzzey <[email protected]>
commit c9723750a699c3bd465493ac2be8992b72ccb105 upstream.
On my platform (i.MX53) bus access sometimes fails with
w1_search: max_slave_count 64 reached, will continue next search.
The reason is the use of jiffies to implement a 200us timeout in mxc_w1_ds2_touch_bit().
On some platforms the jiffies timer resolution is insufficient for this.
Fix by replacing jiffies by ktime_get().
For consistency apply the same change to the other use of jiffies in mxc_w1_ds2_reset_bus().
--- drivers/w1/masters/mxc_w1.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/w1/masters/mxc_w1.c +7 -7
@@ -15,7 +15,7 @@#include <linux/clk.h> #include <linux/delay.h> #include <linux/io.h> -#include <linux/jiffies.h> +#include <linux/ktime.h> #include <linux/module.h> #include <linux/mod_devicetable.h> #include <linux/platform_device.h>
@@ -48,12 +48,12 @@ struct mxc_w1_device {static u8 mxc_w1_ds2_reset_bus (void *data) {struct mxc_w1_device *dev = data; -unsigned long timeout; +ktime_t timeout;writeb (MXC_W1_CONTROL_RPP , dev->regs +MXC_W1_CONTROL );/* Wait for reset sequence 511+512us, use 1500us for sure */ - timeout = jiffies +usecs_to_jiffies (1500 ); + timeout =ktime_add_us (ktime_get (),1500 );udelay (511 +512 );
@@ -63,7 +63,7 @@ static u8 mxc_w1_ds2_reset_bus(void *dat/* PST bit is valid after the RPP bit is self-cleared */ if (!(ctrl &MXC_W1_CONTROL_RPP ))return !(ctrl &MXC_W1_CONTROL_PST ); - }while (time_is_after_jiffies (timeout)); + }while (ktime_before (ktime_get (), timeout));return 1 ; }
@@ -76,12 +76,12 @@ static u8 mxc_w1_ds2_reset_bus(void *datstatic u8 mxc_w1_ds2_touch_bit (void *data,u8 bit) {struct mxc_w1_device *dev = data; -unsigned long timeout; +ktime_t timeout;writeb (MXC_W1_CONTROL_WR (bit), dev->regs +MXC_W1_CONTROL );/* Wait for read/write bit (60us, Max 120us), use 200us for sure */ - timeout = jiffies +usecs_to_jiffies (200 ); + timeout =ktime_add_us (ktime_get (),200 );udelay (60 );
@@ -91,7 +91,7 @@ static u8 mxc_w1_ds2_touch_bit(void *dat/* RDST bit is valid after the WR1/RD bit is self-cleared */ if (!(ctrl &MXC_W1_CONTROL_WR (bit)))return !!(ctrl &MXC_W1_CONTROL_RDST ); - }while (time_is_after_jiffies (timeout)); + }while (ktime_before (ktime_get (), timeout));return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 126/191 scsi: mptfusion: Fix null pointer dereferences in mptscsih_remove()
In reply to Greg Kroah-HartmanFrom: Helge Deller <[email protected]>
commit 2f4843b172c2c0360ee7792ad98025fae7baefde upstream.
The mptscsih_remove() function triggers a kernel oops if the Scsi_Host pointer (ioc->sh) is NULL, as can be seen in this syslog:
ioc0: LSI53C1030 B2: Capabilities={Initiator,Target}
Begin: Waiting for root file system ...
scsi host2: error handler thread failed to spawn, error = -4
mptspi: ioc0: WARNING - Unable to register controller with SCSI subsystem
Backtrace: [<000000001045b7cc>] mptspi_probe+0x248/0x3d0 [mptspi] [<0000000040946470>] pci_device_probe+0x1ac/0x2d8 [<0000000040add668>] really_probe+0x1bc/0x988 [<0000000040ade704>] driver_probe_device+0x160/0x218 [<0000000040adee24>] device_driver_attach+0x160/0x188 [<0000000040adef90>] __driver_attach+0x144/0x320 [<0000000040ad7c78>] bus_for_each_dev+0xd4/0x158 [<0000000040adc138>] driver_attach+0x4c/0x80 [<0000000040adb3ec>] bus_add_driver+0x3e0/0x498 [<0000000040ae0130>] driver_register+0xf4/0x298 [<00000000409450c4>] __pci_register_driver+0x78/0xa8 [<000000000007d248>] mptspi_init+0x18c/0x1c4 [mptspi]
This patch adds the necessary NULL-pointer checks. Successfully tested on a HP C8000 parisc workstation with buggy SCSI drives.
--- drivers/message/fusion/mptscsih.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)
drivers/message/fusion/mptscsih.c +8 -5
@@ -1174,8 +1174,10 @@ mptscsih_remove(struct pci_dev *pdev)MPT_SCSI_HOST *hd;int sz1; -if ((hd =shost_priv (host)) ==NULL ) -return ; +if (host ==NULL ) + hd =NULL ; +else + hd =shost_priv (host);mptscsih_shutdown (pdev);
@@ -1191,14 +1193,15 @@ mptscsih_remove(struct pci_dev *pdev)"Free'd ScsiLookup (%d) memory\n" , ioc->name, sz1)); -kfree (hd->info_kbuf); +if (hd) +kfree (hd->info_kbuf);/* NULL the Scsi_Host pointer */ ioc->sh =NULL ; -scsi_host_put (host); - +if (host) +scsi_host_put (host);mpt_detach (pdev); }
Greg Kroah-Hartman
maintainerPATCH 127/191 scsi: qla2xxx: Fix crash on session cleanup with unload
In reply to Greg Kroah-HartmanFrom: Quinn Tran <[email protected]>
commit 50457dab670f396557e60c07f086358460876353 upstream.
On unload, session cleanup prematurely gave the signal for driver unload path to advance.
--- drivers/scsi/qla2xxx/qla_target.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/scsi/qla2xxx/qla_target.c +7 -6
@@ -1228,14 +1228,15 @@ void qlt_schedule_sess_for_deletion(strucase DSC_DELETE_PEND :return ;case DSC_DELETED : -if (tgt && tgt->tgt_stop && (tgt->sess_count ==0 )) -wake_up_all (&tgt->waitQ); -if (sess->vha->fcport_count ==0 ) -wake_up_all (&sess->vha->fcport_waitQ); -if (!sess->plogi_link[QLT_PLOGI_LINK_SAME_WWN ] && - !sess->plogi_link[QLT_PLOGI_LINK_CONFLICT ]) + !sess->plogi_link[QLT_PLOGI_LINK_CONFLICT ]) { +if (tgt && tgt->tgt_stop && tgt->sess_count ==0 ) +wake_up_all (&tgt->waitQ); + +if (sess->vha->fcport_count ==0 ) +wake_up_all (&sess->vha->fcport_waitQ);return ; + }break ;case DSC_UPD_FCPORT :/*
Greg Kroah-Hartman
maintainerPATCH 128/191 btrfs: qgroup: fix wrong qgroup metadata reserve for delayed inode
In reply to Greg Kroah-HartmanFrom: Qu Wenruo <[email protected]>
commit b4c5d8fdfff3e2b6c4fa4a5043e8946dff500f8c upstream.
For delayed inode facility, qgroup metadata is reserved for it, and later freed.
However we're freeing more bytes than we reserved. In btrfs_delayed_inode_reserve_metadata():
num_bytes = btrfs_calc_metadata_size(fs_info, 1);
...
ret = btrfs_qgroup_reserve_meta_prealloc(root,
fs_info->nodesize, true);
...
if (!ret) {
node->bytes_reserved = num_bytes;But in btrfs_delayed_inode_release_metadata():
if (qgroup_free) btrfs_qgroup_free_meta_prealloc(node->root, node->bytes_reserved); else btrfs_qgroup_convert_reserved_meta(node->root, node->bytes_reserved);
This means, we're always releasing more qgroup metadata rsv than we have reserved.
This won't trigger selftest warning, as btrfs qgroup metadata rsv has extra protection against cases like quota enabled half-way.
But we still need to fix this problem any way.
This patch will use the same num_bytes for qgroup metadata rsv so we could handle it correctly.
--- fs/btrfs/delayed-inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
fs/btrfs/delayed-inode.c +1 -2
@@ -620,8 +620,7 @@ static int btrfs_delayed_inode_reserve_m*/ if (!src_rsv || (!trans->bytes_reserved && src_rsv->type !=BTRFS_BLOCK_RSV_DELALLOC )) { - ret =btrfs_qgroup_reserve_meta_prealloc (root, - fs_info->nodesize, true); + ret =btrfs_qgroup_reserve_meta_prealloc (root, num_bytes, true);if (ret <0 )return ret; ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes,
Greg Kroah-Hartman
maintainerPATCH 129/191 btrfs: improve device scanning messages
In reply to Greg Kroah-HartmanFrom: Anand Jain <[email protected]>
commit 79dae17d8d44b2d15779e332180080af45df5352 upstream.
Systems booting without the initramfs seems to scan an unusual kind of device path (/dev/root). And at a later time, the device is updated to the correct path. We generally print the process name and PID of the process scanning the device but we don't capture the same information if the device path is rescanned with a different pathname.
The current message is too long, so drop the unnecessary UUID and add process name and PID.
While at this also update the duplicate device warning to include the process name and PID so the messages are consistent
CC: [email protected] # 4.19+
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=89721
--- fs/btrfs/volumes.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
fs/btrfs/volumes.c +8 -6
@@ -857,16 +857,18 @@ static noinline struct btrfs_device *devbdput (path_bdev);mutex_unlock (&fs_devices->device_list_mutex);btrfs_warn_in_rcu (device->fs_info, -"duplicate device fsid:devid for %pU:%llu old:%s new:%s" , - disk_super->fsid, devid, -rcu_str_deref (device->name), path); +"duplicate device %s devid %llu generation %llu scanned by %s (%d)" , + path, devid, found_transid, + current->comm, +task_pid_nr (current));return ERR_PTR (-EEXIST ); }bdput (path_bdev);btrfs_info_in_rcu (device->fs_info, -"device fsid %pU devid %llu moved old:%s new:%s" , - disk_super->fsid, devid, -rcu_str_deref (device->name), path); +"devid %llu device path %s changed to %s scanned by %s (%d)" , + devid,rcu_str_deref (device->name), + path, current->comm, +task_pid_nr (current)); } name =rcu_string_strdup (path,GFP_NOFS );
Greg Kroah-Hartman
maintainerPATCH 130/191 btrfs: reschedule if necessary when logging directory items
In reply to Greg Kroah-HartmanFrom: Filipe Manana <[email protected]>
commit bb56f02f26fe23798edb1b2175707419b28c752a upstream.
Logging directories with many entries can take a significant amount of time, and in some cases monopolize a cpu/core for a long time if the logging task doesn't happen to block often enough.
Johannes and Lu Fengqi reported test case generic/041 triggering a soft lockup when the kernel has CONFIG_SOFTLOCKUP_DETECTOR=y. For this test case we log an inode with 3002 hard links, and because the test removed one hard link before fsyncing the file, the inode logging causes the parent directory do be logged as well, which has 6004 directory items to log (3002 BTRFS_DIR_ITEM_KEY items plus 3002 BTRFS_DIR_INDEX_KEY items), so it can take a significant amount of time and trigger the soft lockup.
So just make tree-log.c:log_dir_items() reschedule when necessary, releasing the current search path before doing so and then resume from where it was before the reschedule.
The stack trace produced when the soft lockup happens is the following:
[10480.277653] watchdog: BUG: soft lockup - CPU#2 stuck for 22s! [xfs_io:28172] [10480.279418] Modules linked in: dm_thin_pool dm_persistent_data (...) [10480.284915] irq event stamp: 29646366 [10480.285987] hardirqs last enabled at (29646365): [<ffffffff85249b66>] __slab_alloc.constprop.0+0x56/0x60 [10480.288482] hardirqs last disabled at (29646366): [<ffffffff8579b00d>] irqentry_enter+0x1d/0x50 [10480.290856] softirqs last enabled at (4612): [<ffffffff85a00323>] __do_softirq+0x323/0x56c [10480.293615] softirqs last disabled at (4483): [<ffffffff85800dbf>] asm_call_on_stack+0xf/0x20 [10480.296428] CPU: 2 PID: 28172 Comm: xfs_io Not tainted 5.9.0-rc4-default+ #1248 [10480.298948] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba527-rebuilt.opensuse.org 04/01/2014 [10480.302455] RIP: 0010:__slab_alloc.constprop.0+0x19/0x60 [10480.304151] Code: 86 e8 31 75 21 00 66 66 2e 0f 1f 84 00 00 00 (...) [10480.309558] RSP: 0018:ffffadbe09397a58 EFLAGS: 00000282 [10480.311179] RAX: ffff8a495ab92840 RBX: 0000000000000282 RCX: 0000000000000006 [10480.313242] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff85249b66 [10480.315260] RBP: ffff8a497d04b740 R08: 0000000000000001 R09: 0000000000000001 [10480.317229] R10: ffff8a497d044800 R11: ffff8a495ab93c40 R12: 0000000000000000 [10480.319169] R13: 0000000000000000 R14: 0000000000000c40 R15: ffffffffc01daf70 [10480.321104] FS: 00007fa1dc5c0e40(0000) GS:ffff8a497da00000(0000) knlGS:0000000000000000 [10480.323559] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [10480.325235] CR2: 00007fa1dc5befb8 CR3: 0000000004f8a006 CR4: 0000000000170ea0 [10480.327259] Call Trace: [10480.328286] ? overwrite_item+0x1f0/0x5a0 [btrfs] [10480.329784] __kmalloc+0x831/0xa20 [10480.331009] ? btrfs_get_32+0xb0/0x1d0 [btrfs] [10480.332464] overwrite_item+0x1f0/0x5a0 [btrfs] [10480.333948] log_dir_items+0x2ee/0x570 [btrfs] [10480.335413] log_directory_changes+0x82/0xd0 [btrfs] [10480.336926] btrfs_log_inode+0xc9b/0xda0 [btrfs] [10480.338374] ? init_once+0x20/0x20 [btrfs] [10480.339711] btrfs_log_inode_parent+0x8d3/0xd10 [btrfs] [10480.341257] ? dget_parent+0x97/0x2e0 [10480.342480] btrfs_log_dentry_safe+0x3a/0x50 [btrfs] [10480.343977] btrfs_sync_file+0x24b/0x5e0 [btrfs] [10480.345381] do_fsync+0x38/0x70 [10480.346483] __x64_sys_fsync+0x10/0x20 [10480.347703] do_syscall_64+0x2d/0x70 [10480.348891] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [10480.350444] RIP: 0033:0x7fa1dc80970b [10480.351642] Code: 0f 05 48 3d 00 f0 ff ff 77 45 c3 0f 1f 40 00 48 (...) [10480.356952] RSP: 002b:00007fffb3d081d0 EFLAGS: 00000293 ORIG_RAX: 000000000000004a [10480.359458] RAX: ffffffffffffffda RBX: 0000562d93d45e40 RCX: 00007fa1dc80970b [10480.361426] RDX: 0000562d93d44ab0 RSI: 0000562d93d45e60 RDI: 0000000000000003 [10480.363367] RBP: 0000000000000001 R08: 0000000000000000 R09: 00007fa1dc7b2a40 [10480.365317] R10: 0000562d93d0e366 R11: 0000000000000293 R12: 0000000000000001 [10480.367299] R13: 0000562d93d45290 R14: 0000562d93d45e40 R15: 0000562d93d45e60
--- fs/btrfs/tree-log.c | 8 ++++++++ 1 file changed, 8 insertions(+)
fs/btrfs/tree-log.c +8 -0
@@ -3589,6 +3589,7 @@ static noinline int log_dir_items(struct* search and this search we'll not find the key again and can just * bail. */ +search: ret =btrfs_search_slot (NULL , root, &min_key, path,0 ,0 );if (ret !=0 ) goto done;
@@ -3608,6 +3609,13 @@ static noinline int log_dir_items(structif (min_key.objectid != ino || min_key.type != key_type) goto done; + +if (need_resched ()) { +btrfs_release_path (path); +cond_resched (); + goto search; + } + ret =overwrite_item (trans, log, dst_path, src, i, &min_key);if (ret) {
Greg Kroah-Hartman
maintainerPATCH 131/191 btrfs: send, recompute reference path after orphanization of a directory
In reply to Greg Kroah-HartmanFrom: Filipe Manana <[email protected]>
commit 9c2b4e0347067396ceb3ae929d6888c81d610259 upstream.
During an incremental send, when an inode has multiple new references we might end up emitting rename operations for orphanizations that have a source path that is no longer valid due to a previous orphanization of some directory inode. This causes the receiver to fail since it tries to rename a path that does not exists.
Example reproducer:
$ cat reproducer.sh #!/bin/bash
mkfs.btrfs -f /dev/sdi >/dev/null
mount /dev/sdi /mnt/sdi
touch /mnt/sdi/f1
touch /mnt/sdi/f2
mkdir /mnt/sdi/d1
mkdir /mnt/sdi/d1/d2
# Filesystem looks like:
#
# . (ino 256) # |----- f1 (ino 257) # |----- f2 (ino 258) # |----- d1/ (ino 259) # |----- d2/ (ino 260)
btrfs subvolume snapshot -r /mnt/sdi /mnt/sdi/snap1 btrfs send -f /tmp/snap1.send /mnt/sdi/snap1
# Now do a series of changes such that:
#
# *) inode 258 has one new hardlink and the previous name changed #
# *) both names conflict with the old names of two other inodes:
#
# 1) the new name "d1" conflicts with the old name of inode 259,
# under directory inode 256 (root)
#
# 2) the new name "d2" conflicts with the old name of inode 260
# under directory inode 259
#
# *) inodes 259 and 260 now have the old names of inode 258 #
# *) inode 257 is now located under inode 260 - an inode with a number
# smaller than the inode (258) for which we created a second hard
# link and swapped its names with inodes 259 and 260
#
ln /mnt/sdi/f2 /mnt/sdi/d1/f2_link
mv /mnt/sdi/f1 /mnt/sdi/d1/d2/f1
# Swap d1 and f2.
mv /mnt/sdi/d1 /mnt/sdi/tmp
mv /mnt/sdi/f2 /mnt/sdi/d1
mv /mnt/sdi/tmp /mnt/sdi/f2
# Swap d2 and f2_link
mv /mnt/sdi/f2/d2 /mnt/sdi/tmp
mv /mnt/sdi/f2/f2_link /mnt/sdi/f2/d2
mv /mnt/sdi/tmp /mnt/sdi/f2/f2_link
# Filesystem now looks like:
#
# . (ino 256) # |----- d1 (ino 258) # |----- f2/ (ino 259) # |----- f2_link/ (ino 260) # | |----- f1 (ino 257) # | # |----- d2 (ino 258)
btrfs subvolume snapshot -r /mnt/sdi /mnt/sdi/snap2 btrfs send -f /tmp/snap2.send -p /mnt/sdi/snap1 /mnt/sdi/snap2
mkfs.btrfs -f /dev/sdj >/dev/null
mount /dev/sdj /mnt/sdj
btrfs receive -f /tmp/snap1.send /mnt/sdj
btrfs receive -f /tmp/snap2.send /mnt/sdj
umount /mnt/sdi
umount /mnt/sdj
When executed the receive of the incremental stream fails:
$ ./reproducer.sh Create a readonly snapshot of '/mnt/sdi' in '/mnt/sdi/snap1' At subvol /mnt/sdi/snap1 Create a readonly snapshot of '/mnt/sdi' in '/mnt/sdi/snap2' At subvol /mnt/sdi/snap2 At subvol snap1 At snapshot snap2 ERROR: rename d1/d2 -> o260-6-0 failed: No such file or directory
This happens because:
1) When processing inode 257 we end up computing the name for inode 259 because it is an ancestor in the send snapshot, and at that point it still has its old name, "d1", from the parent snapshot because inode 259 was not yet processed. We then cache that name, which is valid until we start processing inode 259 (or set the progress to 260 after processing its references);
2) Later we start processing inode 258 and collecting all its new references into the list sctx->new_refs. The first reference in the list happens to be the reference for name "d1" while the reference for name "d2" is next (the last element of the list).
We compute the full path "d1/d2" for this second reference and store it in the reference (its ->full_path member). The path used for the new parent directory was "d1" and not "f2" because inode 259, the new parent, was not yet processed;
3) When we start processing the new references at process_recorded_refs() we start with the first reference in the list, for the new name "d1". Because there is a conflicting inode that was not yet processed, which is directory inode 259, we orphanize it, renaming it from "d1" to "o259-6-0";
4) Then we start processing the new reference for name "d2", and we realize it conflicts with the reference of inode 260 in the parent snapshot. So we issue an orphanization operation for inode 260 by emitting a rename operation with a destination path of "o260-6-0" and a source path of "d1/d2" - this source path is the value we stored in the reference earlier at step 2), corresponding to the ->full_path member of the reference, however that path is no longer valid due to the orphanization of the directory inode 259 in step 3). This makes the receiver fail since the path does not exists, it should have been "o259-6-0/d2".
Fix this by recomputing the full path of a reference before emitting an orphanization if we previously orphanized any directory, since that directory could be a parent in the new path. This is a rare scenario so keeping it simple and not checking if that previously orphanized directory is in fact an ancestor of the inode we are trying to orphanize.
A test case for fstests follows soon.
--- fs/btrfs/send.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 72 insertions(+)
fs/btrfs/send.c +72 -0
@@ -3811,6 +3811,72 @@ static int update_ref_path(struct send_c }/* +* When processing the new references for an inode we may orphanize an existing +* directory inode because its old name conflicts with one of the new references +* of the current inode. Later, when processing another new reference of our +* inode, we might need to orphanize another inode, but the path we have in the +* reference reflects the pre-orphanization name of the directory we previously +* orphanized. For example: +* +* parent snapshot looks like: +* +* . (ino 256) +* |----- f1 (ino 257) +* |----- f2 (ino 258) +* |----- d1/ (ino 259) +* |----- d2/ (ino 260) +* +* send snapshot looks like: +* +* . (ino 256) +* |----- d1 (ino 258) +* |----- f2/ (ino 259) +* |----- f2_link/ (ino 260) +* | |----- f1 (ino 257) +* | +* |----- d2 (ino 258) +* +* When processing inode 257 we compute the name for inode 259 as "d1", and we +* cache it in the name cache. Later when we start processing inode 258, when +* collecting all its new references we set a full path of "d1/d2" for its new +* reference with name "d2". When we start processing the new references we +* start by processing the new reference with name "d1", and this results in +* orphanizing inode 259, since its old reference causes a conflict. Then we +* move on the next new reference, with name "d2", and we find out we must +* orphanize inode 260, as its old reference conflicts with ours - but for the +* orphanization we use a source path corresponding to the path we stored in the +* new reference, which is "d1/d2" and not "o259-6-0/d2" - this makes the +* receiver fail since the path component "d1/" no longer exists, it was renamed +* to "o259-6-0/" when processing the previous new reference. So in this case we +* must recompute the path in the new reference and use it for the new +* orphanization operation. +*/ +static int refresh_ref_path (struct send_ctx *sctx,struct recorded_ref *ref) +{ +char *name; +int ret; + + name =kmemdup (ref->name, ref->name_len,GFP_KERNEL ); +if (!name) +return -ENOMEM ; + +fs_path_reset (ref->full_path); + ret =get_cur_path (sctx, ref->dir, ref->dir_gen, ref->full_path); +if (ret <0 ) + goto out; + + ret =fs_path_add (ref->full_path, name, ref->name_len); +if (ret <0 ) + goto out; + +/* Update the reference's base name pointer. */ +set_ref_path (ref, ref->full_path); +out: +kfree (name); +return ret; +} + +/* * This does all the move/link/unlink/rmdir magic. */ static int process_recorded_refs (struct send_ctx *sctx,int *pending_move)
@@ -3940,6 +4006,12 @@ static int process_recorded_refs(structstruct name_cache_entry *nce;struct waiting_dir_move *wdm; +if (orphanized_dir) { + ret =refresh_ref_path (sctx, cur); +if (ret <0 ) + goto out; + } + ret =orphanize_inode (sctx, ow_inode, ow_gen, cur->full_path);if (ret <0 )
Greg Kroah-Hartman
maintainerPATCH 132/191 btrfs: use kvzalloc() to allocate clone_roots in btrfs_ioctl_send()
In reply to Greg Kroah-HartmanFrom: Denis Efremov <[email protected]>
commit 8eb2fd00153a3a96a19c62ac9c6d48c2efebe5e8 upstream.
btrfs_ioctl_send() used open-coded kvzalloc implementation earlier. The code was accidentally replaced with kzalloc() call [1]. Restore the original code by using kvzalloc() to allocate sctx->clone_roots.
[1] https://patchwork.kernel.org/patch/9757891/#20529627
--- fs/btrfs/send.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/send.c +1 -1
@@ -6859,7 +6859,7 @@ long btrfs_ioctl_send(struct file *mnt_f alloc_size =sizeof (struct clone_root ) * (arg->clone_sources_count +1 ); - sctx->clone_roots =kzalloc (alloc_size,GFP_KERNEL ); + sctx->clone_roots =kvzalloc (alloc_size,GFP_KERNEL );if (!sctx->clone_roots) { ret = -ENOMEM ; goto out;
Greg Kroah-Hartman
maintainerPATCH 133/191 btrfs: cleanup cow block on error
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit 572c83acdcdafeb04e70aa46be1fa539310be20c upstream.
In fstest btrfs/064 a transaction abort in __btrfs_cow_block could lead to a system lockup. It gets stuck trying to write back inodes, and the write back thread was trying to lock an extent buffer:
$ cat /proc/2143497/stack [<0>] __btrfs_tree_lock+0x108/0x250 [<0>] lock_extent_buffer_for_io+0x35e/0x3a0 [<0>] btree_write_cache_pages+0x15a/0x3b0 [<0>] do_writepages+0x28/0xb0 [<0>] __writeback_single_inode+0x54/0x5c0 [<0>] writeback_sb_inodes+0x1e8/0x510 [<0>] wb_writeback+0xcc/0x440 [<0>] wb_workfn+0xd7/0x650 [<0>] process_one_work+0x236/0x560 [<0>] worker_thread+0x55/0x3c0 [<0>] kthread+0x13a/0x150 [<0>] ret_from_fork+0x1f/0x30
This is because we got an error while COWing a block, specifically here
if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
ret = btrfs_reloc_cow_block(trans, root, buf, cow);
if (ret) {
btrfs_abort_transaction(trans, ret);
return ret;
}
}[16402.241552] BTRFS: Transaction aborted (error -2) [16402.242362] WARNING: CPU: 1 PID: 2563188 at fs/btrfs/ctree.c:1074 __btrfs_cow_block+0x376/0x540 [16402.249469] CPU: 1 PID: 2563188 Comm: fsstress Not tainted 5.9.0-rc6+ #8 [16402.249936] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-2.fc32 04/01/2014 [16402.250525] RIP: 0010:__btrfs_cow_block+0x376/0x540 [16402.252417] RSP: 0018:ffff9cca40e578b0 EFLAGS: 00010282 [16402.252787] RAX: 0000000000000025 RBX: 0000000000000002 RCX: ffff9132bbd19388 [16402.253278] RDX: 00000000ffffffd8 RSI: 0000000000000027 RDI: ffff9132bbd19380 [16402.254063] RBP: ffff9132b41a49c0 R08: 0000000000000000 R09: 0000000000000000 [16402.254887] R10: 0000000000000000 R11: ffff91324758b080 R12: ffff91326ef17ce0 [16402.255694] R13: ffff91325fc0f000 R14: ffff91326ef176b0 R15: ffff9132815e2000 [16402.256321] FS: 00007f542c6d7b80(0000) GS:ffff9132bbd00000(0000) knlGS:0000000000000000 [16402.256973] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [16402.257374] CR2: 00007f127b83f250 CR3: 0000000133480002 CR4: 0000000000370ee0 [16402.257867] Call Trace: [16402.258072] btrfs_cow_block+0x109/0x230 [16402.258356] btrfs_search_slot+0x530/0x9d0 [16402.258655] btrfs_lookup_file_extent+0x37/0x40 [16402.259155] __btrfs_drop_extents+0x13c/0xd60 [16402.259628] ? btrfs_block_rsv_migrate+0x4f/0xb0 [16402.259949] btrfs_replace_file_extents+0x190/0x820 [16402.260873] btrfs_clone+0x9ae/0xc00 [16402.261139] btrfs_extent_same_range+0x66/0x90 [16402.261771] btrfs_remap_file_range+0x353/0x3b1 [16402.262333] vfs_dedupe_file_range_one.part.0+0xd5/0x140 [16402.262821] vfs_dedupe_file_range+0x189/0x220 [16402.263150] do_vfs_ioctl+0x552/0x700 [16402.263662] __x64_sys_ioctl+0x62/0xb0 [16402.264023] do_syscall_64+0x33/0x40 [16402.264364] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [16402.264862] RIP: 0033:0x7f542c7d15cb [16402.266901] RSP: 002b:00007ffd35944ea8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [16402.267627] RAX: ffffffffffffffda RBX: 00000000009d1968 RCX: 00007f542c7d15cb [16402.268298] RDX: 00000000009d2490 RSI: 00000000c0189436 RDI: 0000000000000003 [16402.268958] RBP: 00000000009d2520 R08: 0000000000000036 R09: 00000000009d2e64 [16402.269726] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000002 [16402.270659] R13: 000000000001f000 R14: 00000000009d1970 R15: 00000000009d2e80 [16402.271498] irq event stamp: 0 [16402.271846] hardirqs last enabled at (0): [<0000000000000000>] 0x0 [16402.272497] hardirqs last disabled at (0): [<ffffffff910dbf59>] copy_process+0x6b9/0x1ba0 [16402.273343] softirqs last enabled at (0): [<ffffffff910dbf59>] copy_process+0x6b9/0x1ba0 [16402.273905] softirqs last disabled at (0): [<0000000000000000>] 0x0 [16402.274338] ---[ end trace 737874a5a41a8236 ]--- [16402.274669] BTRFS: error (device dm-9) in __btrfs_cow_block:1074: errno=-2 No such entry [16402.276179] BTRFS info (device dm-9): forced readonly [16402.277046] BTRFS: error (device dm-9) in btrfs_replace_file_extents:2723: errno=-2 No such entry [16402.278744] BTRFS: error (device dm-9) in __btrfs_cow_block:1074: errno=-2 No such entry [16402.279968] BTRFS: error (device dm-9) in __btrfs_cow_block:1074: errno=-2 No such entry [16402.280582] BTRFS info (device dm-9): balance: ended with status: -30
The problem here is that as soon as we allocate the new block it is locked and marked dirty in the btree inode. This means that we could attempt to writeback this block and need to lock the extent buffer. However we're not unlocking it here and thus we deadlock.
Fix this by unlocking the cow block if we have any errors inside of __btrfs_cow_block, and also free it so we do not leak it.
--- fs/btrfs/ctree.c | 6 ++++++ 1 file changed, 6 insertions(+)
fs/btrfs/ctree.c +6 -0
@@ -1110,6 +1110,8 @@ static noinline int __btrfs_cow_block(st ret =update_ref_for_cow (trans, root, buf, cow, &last_ref);if (ret) { +btrfs_tree_unlock (cow); +free_extent_buffer (cow);btrfs_abort_transaction (trans, ret);return ret; }
@@ -1117,6 +1119,8 @@ static noinline int __btrfs_cow_block(stif (test_bit (BTRFS_ROOT_REF_COWS , &root->state)) { ret =btrfs_reloc_cow_block (trans, root, buf, cow);if (ret) { +btrfs_tree_unlock (cow); +free_extent_buffer (cow);btrfs_abort_transaction (trans, ret);return ret; }
@@ -1149,6 +1153,8 @@ static noinline int __btrfs_cow_block(stif (last_ref) { ret =tree_mod_log_free_eb (buf);if (ret) { +btrfs_tree_unlock (cow); +free_extent_buffer (cow);btrfs_abort_transaction (trans, ret);return ret; }
Greg Kroah-Hartman
maintainerPATCH 134/191 btrfs: fix use-after-free on readahead extent after failure to create it
In reply to Greg Kroah-HartmanFrom: Filipe Manana <[email protected]>
commit 83bc1560e02e25c6439341352024ebe8488f4fbd upstream.
If we fail to find suitable zones for a new readahead extent, we end up leaving a stale pointer in the global readahead extents radix tree (fs_info->reada_tree), which can trigger the following trace later on:
[13367.696354] BUG: kernel NULL pointer dereference, address: 00000000000000b0 [13367.696802] #PF: supervisor read access in kernel mode [13367.697249] #PF: error_code(0x0000) - not-present page [13367.697721] PGD 0 P4D 0 [13367.698171] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI [13367.698632] CPU: 6 PID: 851214 Comm: btrfs Tainted: G W 5.9.0-rc6-btrfs-next-69 #1 [13367.699100] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [13367.700069] RIP: 0010:__lock_acquire+0x20a/0x3970 [13367.700562] Code: ff 1f 0f b7 c0 48 0f (...) [13367.701609] RSP: 0018:ffffb14448f57790 EFLAGS: 00010046 [13367.702140] RAX: 0000000000000000 RBX: 29b935140c15e8cf RCX: 0000000000000000 [13367.702698] RDX: 0000000000000002 RSI: ffffffffb3d66bd0 RDI: 0000000000000046 [13367.703240] RBP: ffff8a52ba8ac040 R08: 00000c2866ad9288 R09: 0000000000000001 [13367.703783] R10: 0000000000000001 R11: 00000000b66d9b53 R12: ffff8a52ba8ac9b0 [13367.704330] R13: 0000000000000000 R14: ffff8a532b6333e8 R15: 0000000000000000 [13367.704880] FS: 00007fe1df6b5700(0000) GS:ffff8a5376600000(0000) knlGS:0000000000000000 [13367.705438] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [13367.705995] CR2: 00000000000000b0 CR3: 000000022cca8004 CR4: 00000000003706e0 [13367.706565] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [13367.707127] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [13367.707686] Call Trace: [13367.708246] ? ___slab_alloc+0x395/0x740 [13367.708820] ? reada_add_block+0xae/0xee0 [btrfs] [13367.709383] lock_acquire+0xb1/0x480 [13367.709955] ? reada_add_block+0xe0/0xee0 [btrfs] [13367.710537] ? reada_add_block+0xae/0xee0 [btrfs] [13367.711097] ? rcu_read_lock_sched_held+0x5d/0x90 [13367.711659] ? kmem_cache_alloc_trace+0x8d2/0x990 [13367.712221] ? lock_acquired+0x33b/0x470 [13367.712784] _raw_spin_lock+0x34/0x80 [13367.713356] ? reada_add_block+0xe0/0xee0 [btrfs] [13367.713966] reada_add_block+0xe0/0xee0 [btrfs] [13367.714529] ? btrfs_root_node+0x15/0x1f0 [btrfs] [13367.715077] btrfs_reada_add+0x117/0x170 [btrfs] [13367.715620] scrub_stripe+0x21e/0x10d0 [btrfs] [13367.716141] ? kvm_sched_clock_read+0x5/0x10 [13367.716657] ? __lock_acquire+0x41e/0x3970 [13367.717184] ? scrub_chunk+0x60/0x140 [btrfs] [13367.717697] ? find_held_lock+0x32/0x90 [13367.718254] ? scrub_chunk+0x60/0x140 [btrfs] [13367.718773] ? lock_acquired+0x33b/0x470 [13367.719278] ? scrub_chunk+0xcd/0x140 [btrfs] [13367.719786] scrub_chunk+0xcd/0x140 [btrfs] [13367.720291] scrub_enumerate_chunks+0x270/0x5c0 [btrfs] [13367.720787] ? finish_wait+0x90/0x90 [13367.721281] btrfs_scrub_dev+0x1ee/0x620 [btrfs] [13367.721762] ? rcu_read_lock_any_held+0x8e/0xb0 [13367.722235] ? preempt_count_add+0x49/0xa0 [13367.722710] ? __sb_start_write+0x19b/0x290 [13367.723192] btrfs_ioctl+0x7f5/0x36f0 [btrfs] [13367.723660] ? __fget_files+0x101/0x1d0 [13367.724118] ? find_held_lock+0x32/0x90 [13367.724559] ? __fget_files+0x101/0x1d0 [13367.724982] ? __x64_sys_ioctl+0x83/0xb0 [13367.725399] __x64_sys_ioctl+0x83/0xb0 [13367.725802] do_syscall_64+0x33/0x80 [13367.726188] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [13367.726574] RIP: 0033:0x7fe1df7add87 [13367.726948] Code: 00 00 00 48 8b 05 09 91 (...) [13367.727763] RSP: 002b:00007fe1df6b4d48 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [13367.728179] RAX: ffffffffffffffda RBX: 000055ce1fb596a0 RCX: 00007fe1df7add87 [13367.728604] RDX: 000055ce1fb596a0 RSI: 00000000c400941b RDI: 0000000000000003 [13367.729021] RBP: 0000000000000000 R08: 00007fe1df6b5700 R09: 0000000000000000 [13367.729431] R10: 00007fe1df6b5700 R11: 0000000000000246 R12: 00007ffd922b07de [13367.729842] R13: 00007ffd922b07df R14: 00007fe1df6b4e40 R15: 0000000000802000 [13367.730275] Modules linked in: btrfs blake2b_generic xor (...) [13367.732638] CR2: 00000000000000b0 [13367.733166] ---[ end trace d298b6805556acd9 ]---
What happens is the following:
1) At reada_find_extent() we don't find any existing readahead extent for the metadata extent starting at logical address X;
2) So we proceed to create a new one. We then call btrfs_map_block() to get information about which stripes contain extent X;
3) After that we iterate over the stripes and create only one zone for the readahead extent - only one because reada_find_zone() returned NULL for all iterations except for one, either because a memory allocation failed or it couldn't find the block group of the extent (it may have just been deleted);
4) We then add the new readahead extent to the readahead extents radix tree at fs_info->reada_tree;
5) Then we iterate over each zone of the new readahead extent, and find that the device used for that zone no longer exists, because it was removed or it was the source device of a device replace operation. Since this left 'have_zone' set to 0, after finishing the loop we jump to the 'error' label, call kfree() on the new readahead extent and return without removing it from the radix tree at fs_info->reada_tree;
6) Any future call to reada_find_extent() for the logical address X will find the stale pointer in the readahead extents radix tree, increment its reference counter, which can trigger the use-after-free right away or return it to the caller reada_add_block() that results in the use-after-free of the example trace above.
So fix this by making sure we delete the readahead extent from the radix tree if we fail to setup zones for it (when 'have_zone = 0').
--- fs/btrfs/reada.c | 2 ++ 1 file changed, 2 insertions(+)
fs/btrfs/reada.c +2 -0
@@ -442,6 +442,8 @@ static struct reada_extent *reada_find_e } have_zone =1 ; } +if (!have_zone) +radix_tree_delete (&fs_info->reada_tree, index);spin_unlock (&fs_info->reada_lock);btrfs_dev_replace_read_unlock (&fs_info->dev_replace);
Greg Kroah-Hartman
maintainerPATCH 135/191 usb: xhci: Workaround for S3 issue on AMD SNPS 3.0 xHC
In reply to Greg Kroah-HartmanFrom: Sandeep Singh <[email protected]>
commit 2a632815683d2d34df52b701a36fe5ac6654e719 upstream.
On some platform of AMD, S3 fails with HCE and SRE errors. To fix this, need to disable a bit which is enable in sparse controller.
--- drivers/usb/host/xhci-pci.c | 17 +++++++++++++++++ drivers/usb/host/xhci.h | 1 + 2 files changed, 18 insertions(+)
drivers/usb/host/xhci-pci.c +17 -0
@@ -21,6 +21,8 @@#define SSIC_PORT_CFG2_OFFSET 0x30#define PROG_DONE (1 << 30)#define SSIC_PORT_UNUSED (1 << 31) +#define SPARSE_DISABLE_BIT 17 +#define SPARSE_CNTL_ENABLE 0xC12C/* Device for a quirk */ #define PCI_VENDOR_ID_FRESCO_LOGIC 0x1b73
@@ -141,6 +143,9 @@ static void xhci_pci_quirks(struct devic (pdev->device == 0x15e0 || pdev->device == 0x15e1)) xhci->quirks |= XHCI_SNPS_BROKEN_SUSPEND; +if (pdev->vendor ==PCI_VENDOR_ID_AMD && pdev->device ==0x15e5 ) + xhci->quirks |=XHCI_DISABLE_SPARSE ; +if (pdev->vendor ==PCI_VENDOR_ID_AMD ) xhci->quirks |=XHCI_TRUST_TX_LENGTH ;
@@ -441,6 +446,15 @@ static void xhci_pme_quirk(struct usb_hcreadl (reg); } +static void xhci_sparse_control_quirk (struct usb_hcd *hcd) +{ +u32 reg; + + reg =readl (hcd->regs +SPARSE_CNTL_ENABLE ); + reg &= ~BIT (SPARSE_DISABLE_BIT ); +writel (reg, hcd->regs +SPARSE_CNTL_ENABLE ); +} +static int xhci_pci_suspend (struct usb_hcd *hcd,bool do_wakeup) {struct xhci_hcd *xhci =hcd_to_xhci (hcd);
@@ -460,6 +474,9 @@ static int xhci_pci_suspend(struct usb_hif (xhci->quirks &XHCI_SSIC_PORT_UNUSED )xhci_ssic_port_unused_quirk (hcd, true); +if (xhci->quirks &XHCI_DISABLE_SPARSE ) +xhci_sparse_control_quirk (hcd); + ret =xhci_suspend (xhci, do_wakeup);if (ret && (xhci->quirks &XHCI_SSIC_PORT_UNUSED ))xhci_ssic_port_unused_quirk (hcd, false);
drivers/usb/host/xhci.h +1 -0
@@ -1872,6 +1872,7 @@ struct xhci_hcd {#define XHCI_ZERO_64B_REGS BIT_ULL(32)#define XHCI_RESET_PLL_ON_DISCONNECT BIT_ULL(34)#define XHCI_SNPS_BROKEN_SUSPEND BIT_ULL(35) +#define XHCI_DISABLE_SPARSE BIT_ULL(38)unsigned int num_active_eps;unsigned int limit_active_eps;
Greg Kroah-Hartman
maintainerPATCH 136/191 usb: dwc3: ep0: Fix ZLP for OUT ep0 requests
In reply to Greg Kroah-HartmanFrom: Thinh Nguyen <[email protected]>
commit 66706077dc89c66a4777a4c6298273816afb848c upstream.
The current ZLP handling for ep0 requests is only for control IN requests. For OUT direction, DWC3 needs to check and setup for MPS alignment.
Usually, control OUT requests can indicate its transfer size via the wLength field of the control message. So usb_request->zero is usually not needed for OUT direction. To handle ZLP OUT for control endpoint, make sure the TRB is MPS size.
--- drivers/usb/dwc3/ep0.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/usb/dwc3/ep0.c +9 -2
@@ -935,12 +935,16 @@ static void dwc3_ep0_xfer_complete(strucstatic void __dwc3_ep0_do_control_data (struct dwc3 *dwc,struct dwc3_ep *dep,struct dwc3_request *req) { +unsigned int trb_length =0 ;int ret; req->direction = !!dep->number;if (req->request.length ==0 ) { -dwc3_ep0_prepare_one_trb (dep, dwc->ep0_trb_addr,0 , +if (!req->direction) + trb_length = dep->endpoint.maxpacket; + +dwc3_ep0_prepare_one_trb (dep, dwc->bounce_addr, trb_length,DWC3_TRBCTL_CONTROL_DATA , false); ret =dwc3_ep0_start_trans (dep); } else if (!IS_ALIGNED(req->request.length, dep->endpoint.maxpacket)
@@ -987,9 +991,12 @@ static void __dwc3_ep0_do_control_data(s req->trb = &dwc->ep0_trb[dep->trb_enqueue -1 ]; +if (!req->direction) + trb_length = dep->endpoint.maxpacket; +/* Now prepare one extra TRB to align transfer size */ dwc3_ep0_prepare_one_trb (dep, dwc->bounce_addr, -0 ,DWC3_TRBCTL_CONTROL_DATA , + trb_length,DWC3_TRBCTL_CONTROL_DATA , false); ret =dwc3_ep0_start_trans (dep); }else {
Greg Kroah-Hartman
maintainerPATCH 137/191 usb: dwc3: gadget: Check MPS of the request length
In reply to Greg Kroah-HartmanFrom: Thinh Nguyen <[email protected]>
commit ca3df3468eec87f6374662f7de425bc44c3810c1 upstream.
When preparing for SG, not all the entries are prepared at once. When resume, don't use the remaining request length to calculate for MPS alignment. Use the entire request->length to do that.
--- drivers/usb/dwc3/gadget.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/dwc3/gadget.c +2 -2
@@ -1057,6 +1057,8 @@ static void dwc3_prepare_one_trb_sg(strustruct scatterlist *s;int i;unsigned int length = req->request.length; +unsigned int maxp =usb_endpoint_maxp (dep->endpoint.desc); +unsigned int rem = length % maxp;unsigned int remaining = req->request.num_mapped_sgs - req->num_queued_sgs;
@@ -1068,8 +1070,6 @@ static void dwc3_prepare_one_trb_sg(stru length -=sg_dma_len (s);for_each_sg (sg, s, remaining, i) { -unsigned int maxp =usb_endpoint_maxp (dep->endpoint.desc); -unsigned int rem = length % maxp;unsigned int trb_length;unsigned chain = true;
Greg Kroah-Hartman
maintainerPATCH 138/191 usb: dwc3: core: add phy cleanup for probe error handling
In reply to Greg Kroah-HartmanFrom: Li Jun <[email protected]>
commit 03c1fd622f72c7624c81b64fdba4a567ae5ee9cb upstream.
Add the phy cleanup if dwc3 mode init fail, which is the missing part of de-init for dwc3 core init.
--- drivers/usb/dwc3/core.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
drivers/usb/dwc3/core.c +11 -0
@@ -1507,6 +1507,17 @@ static int dwc3_probe(struct platform_de err5:dwc3_event_buffers_cleanup (dwc); + +usb_phy_shutdown (dwc->usb2_phy); +usb_phy_shutdown (dwc->usb3_phy); +phy_exit (dwc->usb2_generic_phy); +phy_exit (dwc->usb3_generic_phy); + +usb_phy_set_suspend (dwc->usb2_phy,1 ); +usb_phy_set_suspend (dwc->usb3_phy,1 ); +phy_power_off (dwc->usb2_generic_phy); +phy_power_off (dwc->usb3_generic_phy); +dwc3_ulpi_exit (dwc); err4:
Greg Kroah-Hartman
maintainerPATCH 139/191 usb: dwc3: core: dont trigger runtime pm when remove driver
In reply to Greg Kroah-HartmanFrom: Li Jun <[email protected]>
commit 266d0493900ac5d6a21cdbe6b1624ed2da94d47a upstream.
No need to trigger runtime pm in driver removal, otherwise if user disable auto suspend via sys file, runtime suspend may be entered, which will call dwc3_core_exit() again and there will be clock disable not balance warning:
[ 2026.820154] xhci-hcd xhci-hcd.0.auto: remove, state 4 [ 2026.825268] usb usb2: USB disconnect, device number 1 [ 2026.831017] xhci-hcd xhci-hcd.0.auto: USB bus 2 deregistered [ 2026.836806] xhci-hcd xhci-hcd.0.auto: remove, state 4 [ 2026.842029] usb usb1: USB disconnect, device number 1 [ 2026.848029] xhci-hcd xhci-hcd.0.auto: USB bus 1 deregistered [ 2026.865889] ------------[ cut here ]------------ [ 2026.870506] usb2_ctrl_root_clk already disabled [ 2026.875082] WARNING: CPU: 0 PID: 731 at drivers/clk/clk.c:958 clk_core_disable+0xa0/0xa8 [ 2026.883170] Modules linked in: dwc3(-) phy_fsl_imx8mq_usb [last unloaded: dwc3] [ 2026.890488] CPU: 0 PID: 731 Comm: rmmod Not tainted 5.8.0-rc7-00280-g9d08cca-dirty #245 [ 2026.898489] Hardware name: NXP i.MX8MQ EVK (DT) [ 2026.903020] pstate: 20000085 (nzCv daIf -PAN -UAO BTYPE=--) [ 2026.908594] pc : clk_core_disable+0xa0/0xa8 [ 2026.912777] lr : clk_core_disable+0xa0/0xa8 [ 2026.916958] sp : ffff8000121b39a0 [ 2026.920271] x29: ffff8000121b39a0 x28: ffff0000b11f3700 [ 2026.925583] x27: 0000000000000000 x26: ffff0000b539c700 [ 2026.930895] x25: 000001d7e44e1232 x24: ffff0000b76fa800 [ 2026.936208] x23: ffff0000b76fa6f8 x22: ffff800008d01040 [ 2026.941520] x21: ffff0000b539ce00 x20: ffff0000b7105000 [ 2026.946832] x19: ffff0000b7105000 x18: 0000000000000010 [ 2026.952144] x17: 0000000000000001 x16: 0000000000000000 [ 2026.957456] x15: ffff0000b11f3b70 x14: ffffffffffffffff [ 2026.962768] x13: ffff8000921b36f7 x12: ffff8000121b36ff [ 2026.968080] x11: ffff8000119e1000 x10: ffff800011bf26d0 [ 2026.973392] x9 : 0000000000000000 x8 : ffff800011bf3000 [ 2026.978704] x7 : ffff800010695d68 x6 : 0000000000000252 [ 2026.984016] x5 : ffff0000bb9881f0 x4 : 0000000000000000 [ 2026.989327] x3 : 0000000000000027 x2 : 0000000000000023 [ 2026.994639] x1 : ac2fa471aa7cab00 x0 : 0000000000000000 [ 2026.999951] Call trace: [ 2027.002401] clk_core_disable+0xa0/0xa8 [ 2027.006238] clk_core_disable_lock+0x20/0x38 [ 2027.010508] clk_disable+0x1c/0x28 [ 2027.013911] clk_bulk_disable+0x34/0x50 [ 2027.017758] dwc3_core_exit+0xec/0x110 [dwc3] [ 2027.022122] dwc3_suspend_common+0x84/0x188 [dwc3] [ 2027.026919] dwc3_runtime_suspend+0x74/0x9c [dwc3] [ 2027.031712] pm_generic_runtime_suspend+0x28/0x40 [ 2027.036419] genpd_runtime_suspend+0xa0/0x258 [ 2027.040777] __rpm_callback+0x88/0x140 [ 2027.044526] rpm_callback+0x20/0x80 [ 2027.048015] rpm_suspend+0xd0/0x418 [ 2027.051503] __pm_runtime_suspend+0x58/0xa0 [ 2027.055693] dwc3_runtime_idle+0x7c/0x90 [dwc3] [ 2027.060224] __rpm_callback+0x88/0x140 [ 2027.063973] rpm_idle+0x78/0x150 [ 2027.067201] __pm_runtime_idle+0x58/0xa0 [ 2027.071130] dwc3_remove+0x64/0xc0 [dwc3] [ 2027.075140] platform_drv_remove+0x28/0x48 [ 2027.079239] device_release_driver_internal+0xf4/0x1c0 [ 2027.084377] driver_detach+0x4c/0xd8 [ 2027.087954] bus_remove_driver+0x54/0xa8 [ 2027.091877] driver_unregister+0x2c/0x58 [ 2027.095799] platform_driver_unregister+0x10/0x18 [ 2027.100509] dwc3_driver_exit+0x14/0x1408 [dwc3] [ 2027.105129] __arm64_sys_delete_module+0x178/0x218 [ 2027.109922] el0_svc_common.constprop.0+0x68/0x160 [ 2027.114714] do_el0_svc+0x20/0x80 [ 2027.118031] el0_sync_handler+0x88/0x190 [ 2027.121953] el0_sync+0x140/0x180 [ 2027.125267] ---[ end trace 027f4f8189958f1f ]--- [ 2027.129976] ------------[ cut here ]------------
--- drivers/usb/dwc3/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/dwc3/core.c +2 -2
@@ -1556,9 +1556,9 @@ static int dwc3_remove(struct platform_ddwc3_core_exit (dwc);dwc3_ulpi_exit (dwc); -pm_runtime_put_sync (&pdev->dev); -pm_runtime_allow (&pdev->dev);pm_runtime_disable (&pdev->dev); +pm_runtime_put_noidle (&pdev->dev); +pm_runtime_set_suspended (&pdev->dev);dwc3_free_event_buffers (dwc);dwc3_free_scratch_buffers (dwc);